Hugging Face's logo

ubergarm
/
MiniMax-M2.7-GGUF

Text Generation
GGUF
imatrix
conversational
minimax_m2
ik_llama.cpp
Model card Files
xet
 Community
10
ubergarm commited on
Commit
5c81bbb
·
1 Parent(s): 9f9d8c8

add some early logs

Browse files
Files changed (2) hide show
  1. logs/convert-MiniMax-M2.7.log +1147 -0
  2. logs/imatrix-MiniMax-M2.7-BF16.log +819 -0
logs/convert-MiniMax-M2.7.log ADDED
@@ -0,0 +1,1147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # mainline llama.cpp master@ff5ef8278
2
+
3
+ numactl -N ${SOCKET} -m ${SOCKET} \
4
+ python \
5
+ convert_hf_to_gguf.py \
6
+ --outtype bf16 \
7
+ --split-max-size 50G \
8
+ --outfile /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/ \
9
+ /mnt/data/models/MiniMaxAI/MiniMax-M2.7/
10
+
11
+ INFO:hf-to-gguf:Loading model: MiniMax-M2.7
12
+ INFO:hf-to-gguf:Model architecture: MiniMaxM2ForCausalLM
13
+ INFO:hf-to-gguf:gguf: loading model weight map from 'model.safetensors.index.json'
14
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00000-of-00130.safetensors'
15
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00001-of-00130.safetensors'
16
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00002-of-00130.safetensors'
17
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00003-of-00130.safetensors'
18
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00004-of-00130.safetensors'
19
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00005-of-00130.safetensors'
20
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00006-of-00130.safetensors'
21
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00007-of-00130.safetensors'
22
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00008-of-00130.safetensors'
23
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00009-of-00130.safetensors'
24
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00010-of-00130.safetensors'
25
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00011-of-00130.safetensors'
26
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00012-of-00130.safetensors'
27
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00013-of-00130.safetensors'
28
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00014-of-00130.safetensors'
29
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00015-of-00130.safetensors'
30
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00016-of-00130.safetensors'
31
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00017-of-00130.safetensors'
32
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00018-of-00130.safetensors'
33
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00019-of-00130.safetensors'
34
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00020-of-00130.safetensors'
35
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00021-of-00130.safetensors'
36
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00022-of-00130.safetensors'
37
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00023-of-00130.safetensors'
38
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00024-of-00130.safetensors'
39
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00025-of-00130.safetensors'
40
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00026-of-00130.safetensors'
41
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00027-of-00130.safetensors'
42
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00028-of-00130.safetensors'
43
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00029-of-00130.safetensors'
44
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00030-of-00130.safetensors'
45
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00031-of-00130.safetensors'
46
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00032-of-00130.safetensors'
47
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00033-of-00130.safetensors'
48
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00034-of-00130.safetensors'
49
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00035-of-00130.safetensors'
50
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00036-of-00130.safetensors'
51
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00037-of-00130.safetensors'
52
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00038-of-00130.safetensors'
53
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00039-of-00130.safetensors'
54
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00040-of-00130.safetensors'
55
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00041-of-00130.safetensors'
56
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00042-of-00130.safetensors'
57
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00043-of-00130.safetensors'
58
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00044-of-00130.safetensors'
59
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00045-of-00130.safetensors'
60
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00046-of-00130.safetensors'
61
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00047-of-00130.safetensors'
62
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00048-of-00130.safetensors'
63
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00049-of-00130.safetensors'
64
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00050-of-00130.safetensors'
65
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00051-of-00130.safetensors'
66
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00052-of-00130.safetensors'
67
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00053-of-00130.safetensors'
68
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00054-of-00130.safetensors'
69
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00055-of-00130.safetensors'
70
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00056-of-00130.safetensors'
71
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00057-of-00130.safetensors'
72
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00058-of-00130.safetensors'
73
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00059-of-00130.safetensors'
74
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00060-of-00130.safetensors'
75
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00061-of-00130.safetensors'
76
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00062-of-00130.safetensors'
77
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00063-of-00130.safetensors'
78
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00064-of-00130.safetensors'
79
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00065-of-00130.safetensors'
80
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00066-of-00130.safetensors'
81
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00067-of-00130.safetensors'
82
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00068-of-00130.safetensors'
83
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00069-of-00130.safetensors'
84
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00070-of-00130.safetensors'
85
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00071-of-00130.safetensors'
86
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00072-of-00130.safetensors'
87
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00073-of-00130.safetensors'
88
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00074-of-00130.safetensors'
89
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00075-of-00130.safetensors'
90
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00076-of-00130.safetensors'
91
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00077-of-00130.safetensors'
92
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00078-of-00130.safetensors'
93
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00079-of-00130.safetensors'
94
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00080-of-00130.safetensors'
95
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00081-of-00130.safetensors'
96
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00082-of-00130.safetensors'
97
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00083-of-00130.safetensors'
98
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00084-of-00130.safetensors'
99
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00085-of-00130.safetensors'
100
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00086-of-00130.safetensors'
101
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00087-of-00130.safetensors'
102
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00088-of-00130.safetensors'
103
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00089-of-00130.safetensors'
104
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00090-of-00130.safetensors'
105
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00091-of-00130.safetensors'
106
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00092-of-00130.safetensors'
107
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00093-of-00130.safetensors'
108
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00094-of-00130.safetensors'
109
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00095-of-00130.safetensors'
110
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00096-of-00130.safetensors'
111
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00097-of-00130.safetensors'
112
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00098-of-00130.safetensors'
113
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00099-of-00130.safetensors'
114
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00100-of-00130.safetensors'
115
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00101-of-00130.safetensors'
116
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00102-of-00130.safetensors'
117
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00103-of-00130.safetensors'
118
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00104-of-00130.safetensors'
119
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00105-of-00130.safetensors'
120
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00106-of-00130.safetensors'
121
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00107-of-00130.safetensors'
122
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00108-of-00130.safetensors'
123
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00109-of-00130.safetensors'
124
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00110-of-00130.safetensors'
125
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00111-of-00130.safetensors'
126
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00112-of-00130.safetensors'
127
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00113-of-00130.safetensors'
128
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00114-of-00130.safetensors'
129
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00115-of-00130.safetensors'
130
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00116-of-00130.safetensors'
131
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00117-of-00130.safetensors'
132
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00118-of-00130.safetensors'
133
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00119-of-00130.safetensors'
134
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00120-of-00130.safetensors'
135
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00121-of-00130.safetensors'
136
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00122-of-00130.safetensors'
137
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00123-of-00130.safetensors'
138
+ INFO:hf-to-gguf:gguf: indexing model part 'model-00124-of-00130.safetensors'
139
+ INFO:gguf.gguf_writer:gguf: This GGUF file is for Little Endian only
140
+ INFO:hf-to-gguf:Exporting model...
141
+ INFO:hf-to-gguf:token_embd.weight, torch.bfloat16 --> BF16, shape = {3072, 200064}
142
+ INFO:hf-to-gguf:blk.0.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
143
+ INFO:hf-to-gguf:blk.0.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
144
+ INFO:hf-to-gguf:blk.0.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
145
+ INFO:hf-to-gguf:blk.0.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
146
+ INFO:hf-to-gguf:blk.0.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
147
+ INFO:hf-to-gguf:blk.0.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
148
+ INFO:hf-to-gguf:blk.0.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
149
+ INFO:hf-to-gguf:blk.0.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
150
+ INFO:hf-to-gguf:blk.0.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
151
+ INFO:hf-to-gguf:blk.0.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
152
+ INFO:hf-to-gguf:blk.0.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
153
+ INFO:hf-to-gguf:blk.0.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
154
+ INFO:hf-to-gguf:blk.0.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
155
+ INFO:hf-to-gguf:blk.1.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
156
+ INFO:hf-to-gguf:blk.1.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
157
+ INFO:hf-to-gguf:blk.1.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
158
+ INFO:hf-to-gguf:blk.1.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
159
+ INFO:hf-to-gguf:blk.1.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
160
+ INFO:hf-to-gguf:blk.1.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
161
+ INFO:hf-to-gguf:blk.1.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
162
+ INFO:hf-to-gguf:blk.1.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
163
+ INFO:hf-to-gguf:blk.1.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
164
+ INFO:hf-to-gguf:blk.1.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
165
+ INFO:hf-to-gguf:blk.1.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
166
+ INFO:hf-to-gguf:blk.1.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
167
+ INFO:hf-to-gguf:blk.1.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
168
+ INFO:hf-to-gguf:blk.2.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
169
+ INFO:hf-to-gguf:blk.2.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
170
+ INFO:hf-to-gguf:blk.2.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
171
+ INFO:hf-to-gguf:blk.2.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
172
+ INFO:hf-to-gguf:blk.2.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
173
+ INFO:hf-to-gguf:blk.2.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
174
+ INFO:hf-to-gguf:blk.2.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
175
+ INFO:hf-to-gguf:blk.2.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
176
+ INFO:hf-to-gguf:blk.2.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
177
+ INFO:hf-to-gguf:blk.2.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
178
+ INFO:hf-to-gguf:blk.2.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
179
+ INFO:hf-to-gguf:blk.2.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
180
+ INFO:hf-to-gguf:blk.2.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
181
+ INFO:hf-to-gguf:blk.3.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
182
+ INFO:hf-to-gguf:blk.3.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
183
+ INFO:hf-to-gguf:blk.3.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
184
+ INFO:hf-to-gguf:blk.3.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
185
+ INFO:hf-to-gguf:blk.3.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
186
+ INFO:hf-to-gguf:blk.3.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
187
+ INFO:hf-to-gguf:blk.3.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
188
+ INFO:hf-to-gguf:blk.3.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
189
+ INFO:hf-to-gguf:blk.3.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
190
+ INFO:hf-to-gguf:blk.3.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
191
+ INFO:hf-to-gguf:blk.3.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
192
+ INFO:hf-to-gguf:blk.3.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
193
+ INFO:hf-to-gguf:blk.3.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
194
+ INFO:hf-to-gguf:blk.4.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
195
+ INFO:hf-to-gguf:blk.4.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
196
+ INFO:hf-to-gguf:blk.4.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
197
+ INFO:hf-to-gguf:blk.4.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
198
+ INFO:hf-to-gguf:blk.4.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
199
+ INFO:hf-to-gguf:blk.4.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
200
+ INFO:hf-to-gguf:blk.4.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
201
+ INFO:hf-to-gguf:blk.4.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
202
+ INFO:hf-to-gguf:blk.4.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
203
+ INFO:hf-to-gguf:blk.4.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
204
+ INFO:hf-to-gguf:blk.4.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
205
+ INFO:hf-to-gguf:blk.4.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
206
+ INFO:hf-to-gguf:blk.4.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
207
+ INFO:hf-to-gguf:blk.5.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
208
+ INFO:hf-to-gguf:blk.5.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
209
+ INFO:hf-to-gguf:blk.5.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
210
+ INFO:hf-to-gguf:blk.5.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
211
+ INFO:hf-to-gguf:blk.5.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
212
+ INFO:hf-to-gguf:blk.5.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
213
+ INFO:hf-to-gguf:blk.5.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
214
+ INFO:hf-to-gguf:blk.5.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
215
+ INFO:hf-to-gguf:blk.5.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
216
+ INFO:hf-to-gguf:blk.5.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
217
+ INFO:hf-to-gguf:blk.5.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
218
+ INFO:hf-to-gguf:blk.5.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
219
+ INFO:hf-to-gguf:blk.5.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
220
+ INFO:hf-to-gguf:blk.6.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
221
+ INFO:hf-to-gguf:blk.6.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
222
+ INFO:hf-to-gguf:blk.6.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
223
+ INFO:hf-to-gguf:blk.6.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
224
+ INFO:hf-to-gguf:blk.6.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
225
+ INFO:hf-to-gguf:blk.6.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
226
+ INFO:hf-to-gguf:blk.6.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
227
+ INFO:hf-to-gguf:blk.6.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
228
+ INFO:hf-to-gguf:blk.6.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
229
+ INFO:hf-to-gguf:blk.6.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
230
+ INFO:hf-to-gguf:blk.6.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
231
+ INFO:hf-to-gguf:blk.6.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
232
+ INFO:hf-to-gguf:blk.6.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
233
+ INFO:hf-to-gguf:blk.7.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
234
+ INFO:hf-to-gguf:blk.7.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
235
+ INFO:hf-to-gguf:blk.7.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
236
+ INFO:hf-to-gguf:blk.7.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
237
+ INFO:hf-to-gguf:blk.7.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
238
+ INFO:hf-to-gguf:blk.7.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
239
+ INFO:hf-to-gguf:blk.7.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
240
+ INFO:hf-to-gguf:blk.7.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
241
+ INFO:hf-to-gguf:blk.7.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
242
+ INFO:hf-to-gguf:blk.7.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
243
+ INFO:hf-to-gguf:blk.7.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
244
+ INFO:hf-to-gguf:blk.7.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
245
+ INFO:hf-to-gguf:blk.7.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
246
+ INFO:hf-to-gguf:blk.8.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
247
+ INFO:hf-to-gguf:blk.8.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
248
+ INFO:hf-to-gguf:blk.8.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
249
+ INFO:hf-to-gguf:blk.8.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
250
+ INFO:hf-to-gguf:blk.8.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
251
+ INFO:hf-to-gguf:blk.8.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
252
+ INFO:hf-to-gguf:blk.8.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
253
+ INFO:hf-to-gguf:blk.8.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
254
+ INFO:hf-to-gguf:blk.8.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
255
+ INFO:hf-to-gguf:blk.8.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
256
+ INFO:hf-to-gguf:blk.8.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
257
+ INFO:hf-to-gguf:blk.8.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
258
+ INFO:hf-to-gguf:blk.8.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
259
+ INFO:hf-to-gguf:blk.9.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
260
+ INFO:hf-to-gguf:blk.9.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
261
+ INFO:hf-to-gguf:blk.9.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
262
+ INFO:hf-to-gguf:blk.9.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
263
+ INFO:hf-to-gguf:blk.9.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
264
+ INFO:hf-to-gguf:blk.9.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
265
+ INFO:hf-to-gguf:blk.9.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
266
+ INFO:hf-to-gguf:blk.9.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
267
+ INFO:hf-to-gguf:blk.9.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
268
+ INFO:hf-to-gguf:blk.9.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
269
+ INFO:hf-to-gguf:blk.9.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
270
+ INFO:hf-to-gguf:blk.9.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
271
+ INFO:hf-to-gguf:blk.9.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
272
+ INFO:hf-to-gguf:blk.10.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
273
+ INFO:hf-to-gguf:blk.10.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
274
+ INFO:hf-to-gguf:blk.10.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
275
+ INFO:hf-to-gguf:blk.10.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
276
+ INFO:hf-to-gguf:blk.10.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
277
+ INFO:hf-to-gguf:blk.10.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
278
+ INFO:hf-to-gguf:blk.10.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
279
+ INFO:hf-to-gguf:blk.10.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
280
+ INFO:hf-to-gguf:blk.10.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
281
+ INFO:hf-to-gguf:blk.10.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
282
+ INFO:hf-to-gguf:blk.10.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
283
+ INFO:hf-to-gguf:blk.10.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
284
+ INFO:hf-to-gguf:blk.10.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
285
+ INFO:hf-to-gguf:blk.11.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
286
+ INFO:hf-to-gguf:blk.11.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
287
+ INFO:hf-to-gguf:blk.11.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
288
+ INFO:hf-to-gguf:blk.11.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
289
+ INFO:hf-to-gguf:blk.11.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
290
+ INFO:hf-to-gguf:blk.11.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
291
+ INFO:hf-to-gguf:blk.11.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
292
+ INFO:hf-to-gguf:blk.11.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
293
+ INFO:hf-to-gguf:blk.11.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
294
+ INFO:hf-to-gguf:blk.11.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
295
+ INFO:hf-to-gguf:blk.11.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
296
+ INFO:hf-to-gguf:blk.11.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
297
+ INFO:hf-to-gguf:blk.11.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
298
+ INFO:hf-to-gguf:blk.12.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
299
+ INFO:hf-to-gguf:blk.12.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
300
+ INFO:hf-to-gguf:blk.12.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
301
+ INFO:hf-to-gguf:blk.12.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
302
+ INFO:hf-to-gguf:blk.12.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
303
+ INFO:hf-to-gguf:blk.12.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
304
+ INFO:hf-to-gguf:blk.12.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
305
+ INFO:hf-to-gguf:blk.12.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
306
+ INFO:hf-to-gguf:blk.12.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
307
+ INFO:hf-to-gguf:blk.12.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
308
+ INFO:hf-to-gguf:blk.12.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
309
+ INFO:hf-to-gguf:blk.12.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
310
+ INFO:hf-to-gguf:blk.12.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
311
+ INFO:hf-to-gguf:blk.13.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
312
+ INFO:hf-to-gguf:blk.13.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
313
+ INFO:hf-to-gguf:blk.13.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
314
+ INFO:hf-to-gguf:blk.13.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
315
+ INFO:hf-to-gguf:blk.13.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
316
+ INFO:hf-to-gguf:blk.13.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
317
+ INFO:hf-to-gguf:blk.13.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
318
+ INFO:hf-to-gguf:blk.13.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
319
+ INFO:hf-to-gguf:blk.13.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
320
+ INFO:hf-to-gguf:blk.13.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
321
+ INFO:hf-to-gguf:blk.13.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
322
+ INFO:hf-to-gguf:blk.13.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
323
+ INFO:hf-to-gguf:blk.13.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
324
+ INFO:hf-to-gguf:blk.14.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
325
+ INFO:hf-to-gguf:blk.14.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
326
+ INFO:hf-to-gguf:blk.14.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
327
+ INFO:hf-to-gguf:blk.14.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
328
+ INFO:hf-to-gguf:blk.14.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
329
+ INFO:hf-to-gguf:blk.14.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
330
+ INFO:hf-to-gguf:blk.14.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
331
+ INFO:hf-to-gguf:blk.14.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
332
+ INFO:hf-to-gguf:blk.14.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
333
+ INFO:hf-to-gguf:blk.14.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
334
+ INFO:hf-to-gguf:blk.14.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
335
+ INFO:hf-to-gguf:blk.14.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
336
+ INFO:hf-to-gguf:blk.14.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
337
+ INFO:hf-to-gguf:blk.15.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
338
+ INFO:hf-to-gguf:blk.15.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
339
+ INFO:hf-to-gguf:blk.15.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
340
+ INFO:hf-to-gguf:blk.15.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
341
+ INFO:hf-to-gguf:blk.15.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
342
+ INFO:hf-to-gguf:blk.15.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
343
+ INFO:hf-to-gguf:blk.15.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
344
+ INFO:hf-to-gguf:blk.15.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
345
+ INFO:hf-to-gguf:blk.15.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
346
+ INFO:hf-to-gguf:blk.15.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
347
+ INFO:hf-to-gguf:blk.15.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
348
+ INFO:hf-to-gguf:blk.15.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
349
+ INFO:hf-to-gguf:blk.15.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
350
+ INFO:hf-to-gguf:blk.16.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
351
+ INFO:hf-to-gguf:blk.16.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
352
+ INFO:hf-to-gguf:blk.16.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
353
+ INFO:hf-to-gguf:blk.16.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
354
+ INFO:hf-to-gguf:blk.16.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
355
+ INFO:hf-to-gguf:blk.16.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
356
+ INFO:hf-to-gguf:blk.16.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
357
+ INFO:hf-to-gguf:blk.16.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
358
+ INFO:hf-to-gguf:blk.16.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
359
+ INFO:hf-to-gguf:blk.16.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
360
+ INFO:hf-to-gguf:blk.16.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
361
+ INFO:hf-to-gguf:blk.16.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
362
+ INFO:hf-to-gguf:blk.16.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
363
+ INFO:hf-to-gguf:blk.17.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
364
+ INFO:hf-to-gguf:blk.17.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
365
+ INFO:hf-to-gguf:blk.17.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
366
+ INFO:hf-to-gguf:blk.17.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
367
+ INFO:hf-to-gguf:blk.17.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
368
+ INFO:hf-to-gguf:blk.17.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
369
+ INFO:hf-to-gguf:blk.17.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
370
+ INFO:hf-to-gguf:blk.17.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
371
+ INFO:hf-to-gguf:blk.17.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
372
+ INFO:hf-to-gguf:blk.17.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
373
+ INFO:hf-to-gguf:blk.17.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
374
+ INFO:hf-to-gguf:blk.17.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
375
+ INFO:hf-to-gguf:blk.17.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
376
+ INFO:hf-to-gguf:blk.18.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
377
+ INFO:hf-to-gguf:blk.18.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
378
+ INFO:hf-to-gguf:blk.18.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
379
+ INFO:hf-to-gguf:blk.18.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
380
+ INFO:hf-to-gguf:blk.18.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
381
+ INFO:hf-to-gguf:blk.18.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
382
+ INFO:hf-to-gguf:blk.18.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
383
+ INFO:hf-to-gguf:blk.18.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
384
+ INFO:hf-to-gguf:blk.18.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
385
+ INFO:hf-to-gguf:blk.18.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
386
+ INFO:hf-to-gguf:blk.18.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
387
+ INFO:hf-to-gguf:blk.18.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
388
+ INFO:hf-to-gguf:blk.18.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
389
+ INFO:hf-to-gguf:blk.19.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
390
+ INFO:hf-to-gguf:blk.19.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
391
+ INFO:hf-to-gguf:blk.19.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
392
+ INFO:hf-to-gguf:blk.19.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
393
+ INFO:hf-to-gguf:blk.19.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
394
+ INFO:hf-to-gguf:blk.19.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
395
+ INFO:hf-to-gguf:blk.19.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
396
+ INFO:hf-to-gguf:blk.19.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
397
+ INFO:hf-to-gguf:blk.19.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
398
+ INFO:hf-to-gguf:blk.19.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
399
+ INFO:hf-to-gguf:blk.19.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
400
+ INFO:hf-to-gguf:blk.19.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
401
+ INFO:hf-to-gguf:blk.19.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
402
+ INFO:hf-to-gguf:blk.20.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
403
+ INFO:hf-to-gguf:blk.20.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
404
+ INFO:hf-to-gguf:blk.20.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
405
+ INFO:hf-to-gguf:blk.20.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
406
+ INFO:hf-to-gguf:blk.20.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
407
+ INFO:hf-to-gguf:blk.20.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
408
+ INFO:hf-to-gguf:blk.20.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
409
+ INFO:hf-to-gguf:blk.20.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
410
+ INFO:hf-to-gguf:blk.20.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
411
+ INFO:hf-to-gguf:blk.20.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
412
+ INFO:hf-to-gguf:blk.20.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
413
+ INFO:hf-to-gguf:blk.20.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
414
+ INFO:hf-to-gguf:blk.20.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
415
+ INFO:hf-to-gguf:blk.21.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
416
+ INFO:hf-to-gguf:blk.21.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
417
+ INFO:hf-to-gguf:blk.21.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
418
+ INFO:hf-to-gguf:blk.21.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
419
+ INFO:hf-to-gguf:blk.21.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
420
+ INFO:hf-to-gguf:blk.21.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
421
+ INFO:hf-to-gguf:blk.21.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
422
+ INFO:hf-to-gguf:blk.21.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
423
+ INFO:hf-to-gguf:blk.21.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
424
+ INFO:hf-to-gguf:blk.21.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
425
+ INFO:hf-to-gguf:blk.21.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
426
+ INFO:hf-to-gguf:blk.21.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
427
+ INFO:hf-to-gguf:blk.21.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
428
+ INFO:hf-to-gguf:blk.22.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
429
+ INFO:hf-to-gguf:blk.22.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
430
+ INFO:hf-to-gguf:blk.22.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
431
+ INFO:hf-to-gguf:blk.22.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
432
+ INFO:hf-to-gguf:blk.22.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
433
+ INFO:hf-to-gguf:blk.22.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
434
+ INFO:hf-to-gguf:blk.22.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
435
+ INFO:hf-to-gguf:blk.22.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
436
+ INFO:hf-to-gguf:blk.22.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
437
+ INFO:hf-to-gguf:blk.22.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
438
+ INFO:hf-to-gguf:blk.22.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
439
+ INFO:hf-to-gguf:blk.22.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
440
+ INFO:hf-to-gguf:blk.22.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
441
+ INFO:hf-to-gguf:blk.23.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
442
+ INFO:hf-to-gguf:blk.23.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
443
+ INFO:hf-to-gguf:blk.23.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
444
+ INFO:hf-to-gguf:blk.23.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
445
+ INFO:hf-to-gguf:blk.23.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
446
+ INFO:hf-to-gguf:blk.23.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
447
+ INFO:hf-to-gguf:blk.23.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
448
+ INFO:hf-to-gguf:blk.23.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
449
+ INFO:hf-to-gguf:blk.23.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
450
+ INFO:hf-to-gguf:blk.23.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
451
+ INFO:hf-to-gguf:blk.23.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
452
+ INFO:hf-to-gguf:blk.23.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
453
+ INFO:hf-to-gguf:blk.23.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
454
+ INFO:hf-to-gguf:blk.24.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
455
+ INFO:hf-to-gguf:blk.24.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
456
+ INFO:hf-to-gguf:blk.24.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
457
+ INFO:hf-to-gguf:blk.24.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
458
+ INFO:hf-to-gguf:blk.24.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
459
+ INFO:hf-to-gguf:blk.24.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
460
+ INFO:hf-to-gguf:blk.24.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
461
+ INFO:hf-to-gguf:blk.24.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
462
+ INFO:hf-to-gguf:blk.24.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
463
+ INFO:hf-to-gguf:blk.24.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
464
+ INFO:hf-to-gguf:blk.24.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
465
+ INFO:hf-to-gguf:blk.24.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
466
+ INFO:hf-to-gguf:blk.24.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
467
+ INFO:hf-to-gguf:blk.25.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
468
+ INFO:hf-to-gguf:blk.25.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
469
+ INFO:hf-to-gguf:blk.25.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
470
+ INFO:hf-to-gguf:blk.25.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
471
+ INFO:hf-to-gguf:blk.25.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
472
+ INFO:hf-to-gguf:blk.25.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
473
+ INFO:hf-to-gguf:blk.25.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
474
+ INFO:hf-to-gguf:blk.25.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
475
+ INFO:hf-to-gguf:blk.25.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
476
+ INFO:hf-to-gguf:blk.25.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
477
+ INFO:hf-to-gguf:blk.25.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
478
+ INFO:hf-to-gguf:blk.25.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
479
+ INFO:hf-to-gguf:blk.25.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
480
+ INFO:hf-to-gguf:blk.26.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
481
+ INFO:hf-to-gguf:blk.26.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
482
+ INFO:hf-to-gguf:blk.26.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
483
+ INFO:hf-to-gguf:blk.26.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
484
+ INFO:hf-to-gguf:blk.26.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
485
+ INFO:hf-to-gguf:blk.26.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
486
+ INFO:hf-to-gguf:blk.26.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
487
+ INFO:hf-to-gguf:blk.26.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
488
+ INFO:hf-to-gguf:blk.26.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
489
+ INFO:hf-to-gguf:blk.26.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
490
+ INFO:hf-to-gguf:blk.26.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
491
+ INFO:hf-to-gguf:blk.26.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
492
+ INFO:hf-to-gguf:blk.26.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
493
+ INFO:hf-to-gguf:blk.27.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
494
+ INFO:hf-to-gguf:blk.27.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
495
+ INFO:hf-to-gguf:blk.27.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
496
+ INFO:hf-to-gguf:blk.27.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
497
+ INFO:hf-to-gguf:blk.27.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
498
+ INFO:hf-to-gguf:blk.27.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
499
+ INFO:hf-to-gguf:blk.27.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
500
+ INFO:hf-to-gguf:blk.27.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
501
+ INFO:hf-to-gguf:blk.27.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
502
+ INFO:hf-to-gguf:blk.27.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
503
+ INFO:hf-to-gguf:blk.27.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
504
+ INFO:hf-to-gguf:blk.27.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
505
+ INFO:hf-to-gguf:blk.27.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
506
+ INFO:hf-to-gguf:blk.28.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
507
+ INFO:hf-to-gguf:blk.28.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
508
+ INFO:hf-to-gguf:blk.28.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
509
+ INFO:hf-to-gguf:blk.28.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
510
+ INFO:hf-to-gguf:blk.28.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
511
+ INFO:hf-to-gguf:blk.28.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
512
+ INFO:hf-to-gguf:blk.28.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
513
+ INFO:hf-to-gguf:blk.28.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
514
+ INFO:hf-to-gguf:blk.28.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
515
+ INFO:hf-to-gguf:blk.28.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
516
+ INFO:hf-to-gguf:blk.28.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
517
+ INFO:hf-to-gguf:blk.28.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
518
+ INFO:hf-to-gguf:blk.28.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
519
+ INFO:hf-to-gguf:blk.29.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
520
+ INFO:hf-to-gguf:blk.29.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
521
+ INFO:hf-to-gguf:blk.29.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
522
+ INFO:hf-to-gguf:blk.29.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
523
+ INFO:hf-to-gguf:blk.29.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
524
+ INFO:hf-to-gguf:blk.29.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
525
+ INFO:hf-to-gguf:blk.29.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
526
+ INFO:hf-to-gguf:blk.29.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
527
+ INFO:hf-to-gguf:blk.29.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
528
+ INFO:hf-to-gguf:blk.29.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
529
+ INFO:hf-to-gguf:blk.29.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
530
+ INFO:hf-to-gguf:blk.29.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
531
+ INFO:hf-to-gguf:blk.29.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
532
+ INFO:hf-to-gguf:blk.30.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
533
+ INFO:hf-to-gguf:blk.30.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
534
+ INFO:hf-to-gguf:blk.30.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
535
+ INFO:hf-to-gguf:blk.30.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
536
+ INFO:hf-to-gguf:blk.30.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
537
+ INFO:hf-to-gguf:blk.30.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
538
+ INFO:hf-to-gguf:blk.30.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
539
+ INFO:hf-to-gguf:blk.30.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
540
+ INFO:hf-to-gguf:blk.30.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
541
+ INFO:hf-to-gguf:blk.30.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
542
+ INFO:hf-to-gguf:blk.30.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
543
+ INFO:hf-to-gguf:blk.30.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
544
+ INFO:hf-to-gguf:blk.30.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
545
+ INFO:hf-to-gguf:blk.31.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
546
+ INFO:hf-to-gguf:blk.31.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
547
+ INFO:hf-to-gguf:blk.31.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
548
+ INFO:hf-to-gguf:blk.31.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
549
+ INFO:hf-to-gguf:blk.31.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
550
+ INFO:hf-to-gguf:blk.31.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
551
+ INFO:hf-to-gguf:blk.31.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
552
+ INFO:hf-to-gguf:blk.31.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
553
+ INFO:hf-to-gguf:blk.31.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
554
+ INFO:hf-to-gguf:blk.31.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
555
+ INFO:hf-to-gguf:blk.31.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
556
+ INFO:hf-to-gguf:blk.31.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
557
+ INFO:hf-to-gguf:blk.31.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
558
+ INFO:hf-to-gguf:blk.32.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
559
+ INFO:hf-to-gguf:blk.32.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
560
+ INFO:hf-to-gguf:blk.32.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
561
+ INFO:hf-to-gguf:blk.32.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
562
+ INFO:hf-to-gguf:blk.32.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
563
+ INFO:hf-to-gguf:blk.32.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
564
+ INFO:hf-to-gguf:blk.32.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
565
+ INFO:hf-to-gguf:blk.32.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
566
+ INFO:hf-to-gguf:blk.32.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
567
+ INFO:hf-to-gguf:blk.32.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
568
+ INFO:hf-to-gguf:blk.32.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
569
+ INFO:hf-to-gguf:blk.32.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
570
+ INFO:hf-to-gguf:blk.32.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
571
+ INFO:hf-to-gguf:blk.33.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
572
+ INFO:hf-to-gguf:blk.33.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
573
+ INFO:hf-to-gguf:blk.33.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
574
+ INFO:hf-to-gguf:blk.33.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
575
+ INFO:hf-to-gguf:blk.33.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
576
+ INFO:hf-to-gguf:blk.33.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
577
+ INFO:hf-to-gguf:blk.33.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
578
+ INFO:hf-to-gguf:blk.33.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
579
+ INFO:hf-to-gguf:blk.33.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
580
+ INFO:hf-to-gguf:blk.33.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
581
+ INFO:hf-to-gguf:blk.33.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
582
+ INFO:hf-to-gguf:blk.33.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
583
+ INFO:hf-to-gguf:blk.33.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
584
+ INFO:hf-to-gguf:blk.34.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
585
+ INFO:hf-to-gguf:blk.34.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
586
+ INFO:hf-to-gguf:blk.34.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
587
+ INFO:hf-to-gguf:blk.34.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
588
+ INFO:hf-to-gguf:blk.34.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
589
+ INFO:hf-to-gguf:blk.34.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
590
+ INFO:hf-to-gguf:blk.34.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
591
+ INFO:hf-to-gguf:blk.34.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
592
+ INFO:hf-to-gguf:blk.34.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
593
+ INFO:hf-to-gguf:blk.34.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
594
+ INFO:hf-to-gguf:blk.34.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
595
+ INFO:hf-to-gguf:blk.34.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
596
+ INFO:hf-to-gguf:blk.34.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
597
+ INFO:hf-to-gguf:blk.35.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
598
+ INFO:hf-to-gguf:blk.35.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
599
+ INFO:hf-to-gguf:blk.35.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
600
+ INFO:hf-to-gguf:blk.35.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
601
+ INFO:hf-to-gguf:blk.35.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
602
+ INFO:hf-to-gguf:blk.35.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
603
+ INFO:hf-to-gguf:blk.35.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
604
+ INFO:hf-to-gguf:blk.35.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
605
+ INFO:hf-to-gguf:blk.35.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
606
+ INFO:hf-to-gguf:blk.35.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
607
+ INFO:hf-to-gguf:blk.35.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
608
+ INFO:hf-to-gguf:blk.35.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
609
+ INFO:hf-to-gguf:blk.35.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
610
+ INFO:hf-to-gguf:blk.36.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
611
+ INFO:hf-to-gguf:blk.36.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
612
+ INFO:hf-to-gguf:blk.36.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
613
+ INFO:hf-to-gguf:blk.36.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
614
+ INFO:hf-to-gguf:blk.36.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
615
+ INFO:hf-to-gguf:blk.36.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
616
+ INFO:hf-to-gguf:blk.36.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
617
+ INFO:hf-to-gguf:blk.36.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
618
+ INFO:hf-to-gguf:blk.36.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
619
+ INFO:hf-to-gguf:blk.36.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
620
+ INFO:hf-to-gguf:blk.36.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
621
+ INFO:hf-to-gguf:blk.36.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
622
+ INFO:hf-to-gguf:blk.36.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
623
+ INFO:hf-to-gguf:blk.37.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
624
+ INFO:hf-to-gguf:blk.37.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
625
+ INFO:hf-to-gguf:blk.37.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
626
+ INFO:hf-to-gguf:blk.37.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
627
+ INFO:hf-to-gguf:blk.37.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
628
+ INFO:hf-to-gguf:blk.37.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
629
+ INFO:hf-to-gguf:blk.37.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
630
+ INFO:hf-to-gguf:blk.37.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
631
+ INFO:hf-to-gguf:blk.37.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
632
+ INFO:hf-to-gguf:blk.37.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
633
+ INFO:hf-to-gguf:blk.37.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
634
+ INFO:hf-to-gguf:blk.37.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
635
+ INFO:hf-to-gguf:blk.37.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
636
+ INFO:hf-to-gguf:blk.38.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
637
+ INFO:hf-to-gguf:blk.38.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
638
+ INFO:hf-to-gguf:blk.38.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
639
+ INFO:hf-to-gguf:blk.38.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
640
+ INFO:hf-to-gguf:blk.38.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
641
+ INFO:hf-to-gguf:blk.38.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
642
+ INFO:hf-to-gguf:blk.38.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
643
+ INFO:hf-to-gguf:blk.38.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
644
+ INFO:hf-to-gguf:blk.38.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
645
+ INFO:hf-to-gguf:blk.38.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
646
+ INFO:hf-to-gguf:blk.38.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
647
+ INFO:hf-to-gguf:blk.38.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
648
+ INFO:hf-to-gguf:blk.38.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
649
+ INFO:hf-to-gguf:blk.39.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
650
+ INFO:hf-to-gguf:blk.39.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
651
+ INFO:hf-to-gguf:blk.39.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
652
+ INFO:hf-to-gguf:blk.39.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
653
+ INFO:hf-to-gguf:blk.39.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
654
+ INFO:hf-to-gguf:blk.39.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
655
+ INFO:hf-to-gguf:blk.39.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
656
+ INFO:hf-to-gguf:blk.39.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
657
+ INFO:hf-to-gguf:blk.39.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
658
+ INFO:hf-to-gguf:blk.39.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
659
+ INFO:hf-to-gguf:blk.39.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
660
+ INFO:hf-to-gguf:blk.39.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
661
+ INFO:hf-to-gguf:blk.39.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
662
+ INFO:hf-to-gguf:blk.40.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
663
+ INFO:hf-to-gguf:blk.40.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
664
+ INFO:hf-to-gguf:blk.40.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
665
+ INFO:hf-to-gguf:blk.40.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
666
+ INFO:hf-to-gguf:blk.40.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
667
+ INFO:hf-to-gguf:blk.40.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
668
+ INFO:hf-to-gguf:blk.40.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
669
+ INFO:hf-to-gguf:blk.40.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
670
+ INFO:hf-to-gguf:blk.40.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
671
+ INFO:hf-to-gguf:blk.40.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
672
+ INFO:hf-to-gguf:blk.40.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
673
+ INFO:hf-to-gguf:blk.40.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
674
+ INFO:hf-to-gguf:blk.40.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
675
+ INFO:hf-to-gguf:blk.41.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
676
+ INFO:hf-to-gguf:blk.41.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
677
+ INFO:hf-to-gguf:blk.41.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
678
+ INFO:hf-to-gguf:blk.41.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
679
+ INFO:hf-to-gguf:blk.41.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
680
+ INFO:hf-to-gguf:blk.41.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
681
+ INFO:hf-to-gguf:blk.41.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
682
+ INFO:hf-to-gguf:blk.41.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
683
+ INFO:hf-to-gguf:blk.41.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
684
+ INFO:hf-to-gguf:blk.41.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
685
+ INFO:hf-to-gguf:blk.41.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
686
+ INFO:hf-to-gguf:blk.41.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
687
+ INFO:hf-to-gguf:blk.41.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
688
+ INFO:hf-to-gguf:blk.42.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
689
+ INFO:hf-to-gguf:blk.42.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
690
+ INFO:hf-to-gguf:blk.42.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
691
+ INFO:hf-to-gguf:blk.42.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
692
+ INFO:hf-to-gguf:blk.42.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
693
+ INFO:hf-to-gguf:blk.42.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
694
+ INFO:hf-to-gguf:blk.42.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
695
+ INFO:hf-to-gguf:blk.42.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
696
+ INFO:hf-to-gguf:blk.42.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
697
+ INFO:hf-to-gguf:blk.42.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
698
+ INFO:hf-to-gguf:blk.42.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
699
+ INFO:hf-to-gguf:blk.42.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
700
+ INFO:hf-to-gguf:blk.42.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
701
+ INFO:hf-to-gguf:blk.43.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
702
+ INFO:hf-to-gguf:blk.43.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
703
+ INFO:hf-to-gguf:blk.43.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
704
+ INFO:hf-to-gguf:blk.43.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
705
+ INFO:hf-to-gguf:blk.43.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
706
+ INFO:hf-to-gguf:blk.43.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
707
+ INFO:hf-to-gguf:blk.43.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
708
+ INFO:hf-to-gguf:blk.43.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
709
+ INFO:hf-to-gguf:blk.43.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
710
+ INFO:hf-to-gguf:blk.43.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
711
+ INFO:hf-to-gguf:blk.43.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
712
+ INFO:hf-to-gguf:blk.43.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
713
+ INFO:hf-to-gguf:blk.43.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
714
+ INFO:hf-to-gguf:blk.44.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
715
+ INFO:hf-to-gguf:blk.44.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
716
+ INFO:hf-to-gguf:blk.44.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
717
+ INFO:hf-to-gguf:blk.44.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
718
+ INFO:hf-to-gguf:blk.44.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
719
+ INFO:hf-to-gguf:blk.44.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
720
+ INFO:hf-to-gguf:blk.44.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
721
+ INFO:hf-to-gguf:blk.44.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
722
+ INFO:hf-to-gguf:blk.44.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
723
+ INFO:hf-to-gguf:blk.44.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
724
+ INFO:hf-to-gguf:blk.44.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
725
+ INFO:hf-to-gguf:blk.44.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
726
+ INFO:hf-to-gguf:blk.44.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
727
+ INFO:hf-to-gguf:blk.45.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
728
+ INFO:hf-to-gguf:blk.45.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
729
+ INFO:hf-to-gguf:blk.45.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
730
+ INFO:hf-to-gguf:blk.45.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
731
+ INFO:hf-to-gguf:blk.45.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
732
+ INFO:hf-to-gguf:blk.45.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
733
+ INFO:hf-to-gguf:blk.45.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
734
+ INFO:hf-to-gguf:blk.45.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
735
+ INFO:hf-to-gguf:blk.45.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
736
+ INFO:hf-to-gguf:blk.45.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
737
+ INFO:hf-to-gguf:blk.45.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
738
+ INFO:hf-to-gguf:blk.45.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
739
+ INFO:hf-to-gguf:blk.45.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
740
+ INFO:hf-to-gguf:blk.46.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
741
+ INFO:hf-to-gguf:blk.46.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
742
+ INFO:hf-to-gguf:blk.46.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
743
+ INFO:hf-to-gguf:blk.46.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
744
+ INFO:hf-to-gguf:blk.46.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
745
+ INFO:hf-to-gguf:blk.46.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
746
+ INFO:hf-to-gguf:blk.46.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
747
+ INFO:hf-to-gguf:blk.46.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
748
+ INFO:hf-to-gguf:blk.46.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
749
+ INFO:hf-to-gguf:blk.46.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
750
+ INFO:hf-to-gguf:blk.46.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
751
+ INFO:hf-to-gguf:blk.46.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
752
+ INFO:hf-to-gguf:blk.46.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
753
+ INFO:hf-to-gguf:blk.47.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
754
+ INFO:hf-to-gguf:blk.47.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
755
+ INFO:hf-to-gguf:blk.47.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
756
+ INFO:hf-to-gguf:blk.47.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
757
+ INFO:hf-to-gguf:blk.47.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
758
+ INFO:hf-to-gguf:blk.47.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
759
+ INFO:hf-to-gguf:blk.47.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
760
+ INFO:hf-to-gguf:blk.47.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
761
+ INFO:hf-to-gguf:blk.47.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
762
+ INFO:hf-to-gguf:blk.47.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
763
+ INFO:hf-to-gguf:blk.47.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
764
+ INFO:hf-to-gguf:blk.47.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
765
+ INFO:hf-to-gguf:blk.47.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
766
+ INFO:hf-to-gguf:blk.48.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
767
+ INFO:hf-to-gguf:blk.48.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
768
+ INFO:hf-to-gguf:blk.48.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
769
+ INFO:hf-to-gguf:blk.48.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
770
+ INFO:hf-to-gguf:blk.48.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
771
+ INFO:hf-to-gguf:blk.48.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
772
+ INFO:hf-to-gguf:blk.48.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
773
+ INFO:hf-to-gguf:blk.48.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
774
+ INFO:hf-to-gguf:blk.48.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
775
+ INFO:hf-to-gguf:blk.48.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
776
+ INFO:hf-to-gguf:blk.48.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
777
+ INFO:hf-to-gguf:blk.48.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
778
+ INFO:hf-to-gguf:blk.48.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
779
+ INFO:hf-to-gguf:blk.49.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
780
+ INFO:hf-to-gguf:blk.49.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
781
+ INFO:hf-to-gguf:blk.49.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
782
+ INFO:hf-to-gguf:blk.49.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
783
+ INFO:hf-to-gguf:blk.49.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
784
+ INFO:hf-to-gguf:blk.49.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
785
+ INFO:hf-to-gguf:blk.49.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
786
+ INFO:hf-to-gguf:blk.49.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
787
+ INFO:hf-to-gguf:blk.49.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
788
+ INFO:hf-to-gguf:blk.49.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
789
+ INFO:hf-to-gguf:blk.49.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
790
+ INFO:hf-to-gguf:blk.49.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
791
+ INFO:hf-to-gguf:blk.49.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
792
+ INFO:hf-to-gguf:blk.50.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
793
+ INFO:hf-to-gguf:blk.50.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
794
+ INFO:hf-to-gguf:blk.50.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
795
+ INFO:hf-to-gguf:blk.50.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
796
+ INFO:hf-to-gguf:blk.50.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
797
+ INFO:hf-to-gguf:blk.50.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
798
+ INFO:hf-to-gguf:blk.50.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
799
+ INFO:hf-to-gguf:blk.50.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
800
+ INFO:hf-to-gguf:blk.50.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
801
+ INFO:hf-to-gguf:blk.50.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
802
+ INFO:hf-to-gguf:blk.50.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
803
+ INFO:hf-to-gguf:blk.50.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
804
+ INFO:hf-to-gguf:blk.50.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
805
+ INFO:hf-to-gguf:blk.51.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
806
+ INFO:hf-to-gguf:blk.51.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
807
+ INFO:hf-to-gguf:blk.51.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
808
+ INFO:hf-to-gguf:blk.51.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
809
+ INFO:hf-to-gguf:blk.51.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
810
+ INFO:hf-to-gguf:blk.51.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
811
+ INFO:hf-to-gguf:blk.51.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
812
+ INFO:hf-to-gguf:blk.51.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
813
+ INFO:hf-to-gguf:blk.51.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
814
+ INFO:hf-to-gguf:blk.51.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
815
+ INFO:hf-to-gguf:blk.51.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
816
+ INFO:hf-to-gguf:blk.51.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
817
+ INFO:hf-to-gguf:blk.51.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
818
+ INFO:hf-to-gguf:blk.52.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
819
+ INFO:hf-to-gguf:blk.52.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
820
+ INFO:hf-to-gguf:blk.52.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
821
+ INFO:hf-to-gguf:blk.52.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
822
+ INFO:hf-to-gguf:blk.52.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
823
+ INFO:hf-to-gguf:blk.52.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
824
+ INFO:hf-to-gguf:blk.52.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
825
+ INFO:hf-to-gguf:blk.52.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
826
+ INFO:hf-to-gguf:blk.52.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
827
+ INFO:hf-to-gguf:blk.52.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
828
+ INFO:hf-to-gguf:blk.52.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
829
+ INFO:hf-to-gguf:blk.52.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
830
+ INFO:hf-to-gguf:blk.52.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
831
+ INFO:hf-to-gguf:blk.53.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
832
+ INFO:hf-to-gguf:blk.53.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
833
+ INFO:hf-to-gguf:blk.53.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
834
+ INFO:hf-to-gguf:blk.53.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
835
+ INFO:hf-to-gguf:blk.53.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
836
+ INFO:hf-to-gguf:blk.53.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
837
+ INFO:hf-to-gguf:blk.53.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
838
+ INFO:hf-to-gguf:blk.53.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
839
+ INFO:hf-to-gguf:blk.53.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
840
+ INFO:hf-to-gguf:blk.53.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
841
+ INFO:hf-to-gguf:blk.53.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
842
+ INFO:hf-to-gguf:blk.53.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
843
+ INFO:hf-to-gguf:blk.53.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
844
+ INFO:hf-to-gguf:blk.54.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
845
+ INFO:hf-to-gguf:blk.54.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
846
+ INFO:hf-to-gguf:blk.54.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
847
+ INFO:hf-to-gguf:blk.54.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
848
+ INFO:hf-to-gguf:blk.54.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
849
+ INFO:hf-to-gguf:blk.54.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
850
+ INFO:hf-to-gguf:blk.54.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
851
+ INFO:hf-to-gguf:blk.54.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
852
+ INFO:hf-to-gguf:blk.54.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
853
+ INFO:hf-to-gguf:blk.54.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
854
+ INFO:hf-to-gguf:blk.54.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
855
+ INFO:hf-to-gguf:blk.54.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
856
+ INFO:hf-to-gguf:blk.54.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
857
+ INFO:hf-to-gguf:blk.55.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
858
+ INFO:hf-to-gguf:blk.55.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
859
+ INFO:hf-to-gguf:blk.55.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
860
+ INFO:hf-to-gguf:blk.55.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
861
+ INFO:hf-to-gguf:blk.55.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
862
+ INFO:hf-to-gguf:blk.55.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
863
+ INFO:hf-to-gguf:blk.55.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
864
+ INFO:hf-to-gguf:blk.55.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
865
+ INFO:hf-to-gguf:blk.55.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
866
+ INFO:hf-to-gguf:blk.55.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
867
+ INFO:hf-to-gguf:blk.55.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
868
+ INFO:hf-to-gguf:blk.55.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
869
+ INFO:hf-to-gguf:blk.55.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
870
+ INFO:hf-to-gguf:blk.56.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
871
+ INFO:hf-to-gguf:blk.56.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
872
+ INFO:hf-to-gguf:blk.56.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
873
+ INFO:hf-to-gguf:blk.56.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
874
+ INFO:hf-to-gguf:blk.56.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
875
+ INFO:hf-to-gguf:blk.56.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
876
+ INFO:hf-to-gguf:blk.56.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
877
+ INFO:hf-to-gguf:blk.56.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
878
+ INFO:hf-to-gguf:blk.56.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
879
+ INFO:hf-to-gguf:blk.56.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
880
+ INFO:hf-to-gguf:blk.56.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
881
+ INFO:hf-to-gguf:blk.56.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
882
+ INFO:hf-to-gguf:blk.56.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
883
+ INFO:hf-to-gguf:blk.57.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
884
+ INFO:hf-to-gguf:blk.57.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
885
+ INFO:hf-to-gguf:blk.57.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
886
+ INFO:hf-to-gguf:blk.57.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
887
+ INFO:hf-to-gguf:blk.57.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
888
+ INFO:hf-to-gguf:blk.57.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
889
+ INFO:hf-to-gguf:blk.57.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
890
+ INFO:hf-to-gguf:blk.57.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
891
+ INFO:hf-to-gguf:blk.57.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
892
+ INFO:hf-to-gguf:blk.57.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
893
+ INFO:hf-to-gguf:blk.57.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
894
+ INFO:hf-to-gguf:blk.57.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
895
+ INFO:hf-to-gguf:blk.57.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
896
+ INFO:hf-to-gguf:blk.58.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
897
+ INFO:hf-to-gguf:blk.58.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
898
+ INFO:hf-to-gguf:blk.58.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
899
+ INFO:hf-to-gguf:blk.58.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
900
+ INFO:hf-to-gguf:blk.58.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
901
+ INFO:hf-to-gguf:blk.58.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
902
+ INFO:hf-to-gguf:blk.58.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
903
+ INFO:hf-to-gguf:blk.58.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
904
+ INFO:hf-to-gguf:blk.58.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
905
+ INFO:hf-to-gguf:blk.58.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
906
+ INFO:hf-to-gguf:blk.58.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
907
+ INFO:hf-to-gguf:blk.58.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
908
+ INFO:hf-to-gguf:blk.58.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
909
+ INFO:hf-to-gguf:blk.59.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
910
+ INFO:hf-to-gguf:blk.59.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
911
+ INFO:hf-to-gguf:blk.59.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
912
+ INFO:hf-to-gguf:blk.59.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
913
+ INFO:hf-to-gguf:blk.59.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
914
+ INFO:hf-to-gguf:blk.59.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
915
+ INFO:hf-to-gguf:blk.59.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
916
+ INFO:hf-to-gguf:blk.59.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
917
+ INFO:hf-to-gguf:blk.59.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
918
+ INFO:hf-to-gguf:blk.59.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
919
+ INFO:hf-to-gguf:blk.59.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
920
+ INFO:hf-to-gguf:blk.59.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
921
+ INFO:hf-to-gguf:blk.59.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
922
+ INFO:hf-to-gguf:blk.60.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
923
+ INFO:hf-to-gguf:blk.60.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
924
+ INFO:hf-to-gguf:blk.60.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
925
+ INFO:hf-to-gguf:blk.60.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
926
+ INFO:hf-to-gguf:blk.60.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
927
+ INFO:hf-to-gguf:blk.60.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
928
+ INFO:hf-to-gguf:blk.60.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
929
+ INFO:hf-to-gguf:blk.60.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
930
+ INFO:hf-to-gguf:blk.60.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
931
+ INFO:hf-to-gguf:blk.60.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
932
+ INFO:hf-to-gguf:blk.60.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
933
+ INFO:hf-to-gguf:blk.60.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
934
+ INFO:hf-to-gguf:blk.60.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
935
+ INFO:hf-to-gguf:blk.61.exp_probs_b.bias, torch.float32 --> F32, shape = {256}
936
+ INFO:hf-to-gguf:blk.61.ffn_gate_inp.weight, torch.float32 --> F32, shape = {3072, 256}
937
+ INFO:hf-to-gguf:blk.61.attn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
938
+ INFO:hf-to-gguf:blk.61.ffn_norm.weight, torch.bfloat16 --> F32, shape = {3072}
939
+ INFO:hf-to-gguf:blk.61.attn_k_norm.weight, torch.bfloat16 --> F32, shape = {1024}
940
+ INFO:hf-to-gguf:blk.61.attn_k.weight, torch.float32 --> BF16, shape = {3072, 1024}
941
+ INFO:hf-to-gguf:blk.61.attn_output.weight, torch.float32 --> BF16, shape = {6144, 3072}
942
+ INFO:hf-to-gguf:blk.61.attn_q_norm.weight, torch.bfloat16 --> F32, shape = {6144}
943
+ INFO:hf-to-gguf:blk.61.attn_q.weight, torch.float32 --> BF16, shape = {3072, 6144}
944
+ INFO:hf-to-gguf:blk.61.attn_v.weight, torch.float32 --> BF16, shape = {3072, 1024}
945
+ INFO:hf-to-gguf:blk.61.ffn_gate_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
946
+ INFO:hf-to-gguf:blk.61.ffn_down_exps.weight, torch.float32 --> BF16, shape = {1536, 3072, 256}
947
+ INFO:hf-to-gguf:blk.61.ffn_up_exps.weight, torch.float32 --> BF16, shape = {3072, 1536, 256}
948
+ INFO:hf-to-gguf:output.weight, torch.bfloat16 --> BF16, shape = {3072, 200064}
949
+ INFO:hf-to-gguf:output_norm.weight, torch.bfloat16 --> F32, shape = {3072}
950
+ INFO:hf-to-gguf:Set meta model
951
+ INFO:hf-to-gguf:Set model parameters
952
+ INFO:hf-to-gguf:gguf: context length = 196608
953
+ INFO:hf-to-gguf:gguf: embedding length = 3072
954
+ INFO:hf-to-gguf:gguf: feed forward length = 1536
955
+ INFO:hf-to-gguf:gguf: head count = 48
956
+ INFO:hf-to-gguf:gguf: key-value head count = 8
957
+ WARNING:hf-to-gguf:Unknown RoPE type: default
958
+ INFO:hf-to-gguf:gguf: rope scaling type = NONE
959
+ INFO:hf-to-gguf:gguf: rope theta = 5000000
960
+ INFO:hf-to-gguf:gguf: rms norm epsilon = 1e-06
961
+ INFO:hf-to-gguf:gguf: expert count = 256
962
+ INFO:hf-to-gguf:gguf: experts used count = 8
963
+ INFO:hf-to-gguf:gguf: expert score gating function = sigmoid
964
+ INFO:hf-to-gguf:gguf: file type = 32
965
+ INFO:hf-to-gguf:Set model quantization version
966
+ INFO:hf-to-gguf:Set model tokenizer
967
+ INFO:gguf.vocab:Adding 199744 merge(s).
968
+ INFO:gguf.vocab:Setting special token type bos to 200034
969
+ INFO:gguf.vocab:Setting special token type eos to 200020
970
+ INFO:gguf.vocab:Setting special token type unk to 200021
971
+ INFO:gguf.vocab:Setting chat_template to {# ----------‑‑‑ special token variables ‑‑‑---------- #}
972
+ {%- set toolcall_begin_token = '<minimax:tool_call>' -%}
973
+ {%- set toolcall_end_token = '</minimax:tool_call>' -%}
974
+ {#- Tool Rendering Functions ============================================== -#}
975
+ {%- macro render_tool_namespace(namespace_name, tool_list) -%}
976
+ {%- for tool in tool_list -%}
977
+ <tool>{{ tool.function | tojson(ensure_ascii=False) }}</tool>
978
+ {% endfor -%}
979
+ {%- endmacro -%}
980
+ {%- macro visible_text(content) -%}
981
+ {%- if content is string -%}
982
+ {{ content }}
983
+ {%- elif content is iterable and content is not mapping -%}
984
+ {%- for item in content -%}
985
+ {%- if item is mapping and item.type == 'text' -%}
986
+ {{- item.text }}
987
+ {%- elif item is string -%}
988
+ {{- item }}
989
+ {%- endif -%}
990
+ {%- endfor -%}
991
+ {%- else -%}
992
+ {{- content }}
993
+ {%- endif -%}
994
+ {%- endmacro -%}
995
+ {#- System Message Construction ============================================ -#}
996
+ {%- macro build_system_message(system_message) -%}
997
+ {%- if system_message and system_message.content -%}
998
+ {{- visible_text(system_message.content) }}
999
+ {%- else -%}
1000
+ {%- if model_identity is not defined -%}
1001
+ {%- set model_identity = "You are a helpful assistant. Your name is MiniMax-M2.7 and is built by MiniMax." -%}
1002
+ {%- endif -%}
1003
+ {{- model_identity }}
1004
+ {%- endif -%}
1005
+
1006
+ {#- Handle current_date -#}
1007
+ {%- if system_message and system_message.current_date -%}
1008
+ {{- '\n' ~ 'Current date: ' + system_message.current_date }}
1009
+ {%- endif -%}
1010
+ {#- Handle current_location -#}
1011
+ {%- if system_message and system_message.current_location -%}
1012
+ {{- '\n' ~ 'Current location: ' + system_message.current_location }}
1013
+ {%- endif -%}
1014
+ {%- endmacro -%}
1015
+ {#- Main Template Logic ================================================= -#}
1016
+ {#- Extract system message (only first message if it's system) -#}
1017
+ {%- set system_message = none -%}
1018
+ {%- set conversation_messages = messages -%}
1019
+ {%- if messages and messages[0].role == "system" -%}
1020
+ {%- set system_message = messages[0] -%}
1021
+ {%- set conversation_messages = messages[1:] -%}
1022
+ {%- endif -%}
1023
+ {#- Get the last user message turn, for interleved thinking -#}
1024
+ {%- set ns = namespace(last_user_index=-1) %}
1025
+ {% for m in conversation_messages %}
1026
+ {%- if m.role == 'user' %}
1027
+ {% set ns.last_user_index = loop.index0 -%}
1028
+ {%- endif %}
1029
+ {%- endfor %}
1030
+ {#- Render system message -#}
1031
+ {{- ']~!b[' ~ ']~b]system' ~ '\n' }}
1032
+ {{- build_system_message(system_message) }}
1033
+ {#- Render tools if available -#}
1034
+ {%- if tools -%}
1035
+ {{- '\n\n' ~ '# Tools' ~ '\n' ~ 'You may call one or more tools to assist with the user query.\nHere are the tools available in JSONSchema format:' ~ '\n' }}
1036
+ {{- '\n' ~ '<tools>' ~ '\n' }}
1037
+ {{- render_tool_namespace("functions", tools) }}
1038
+ {{- '</tools>' ~ '\n\n' }}
1039
+ {{- 'When making tool calls, use XML format to invoke tools and pass parameters:' ~ '\n' }}
1040
+ {{- '\n' ~ toolcall_begin_token }}
1041
+ <invoke name="tool-name-1">
1042
+ <parameter name="param-key-1">param-value-1</parameter>
1043
+ <parameter name="param-key-2">param-value-2</parameter>
1044
+ ...
1045
+ </invoke>
1046
+ {{- '\n' ~ toolcall_end_token }}
1047
+ {%- endif -%}
1048
+ {{- '[e~[\n' }}
1049
+
1050
+ {#- Render messages -#}
1051
+ {%- set last_tool_call = namespace(name=none) -%}
1052
+ {%- for message in conversation_messages -%}
1053
+ {%- if message.role == 'assistant' -%}
1054
+ {#- Only render reasoning_content if no user message follows -#}
1055
+ {{- ']~b]ai' ~ '\n' }}
1056
+
1057
+ {%- set reasoning_content = '' %}
1058
+ {%- set content = visible_text(message.content) %}
1059
+ {%- if message.reasoning_content is string %}
1060
+ {%- set reasoning_content = message.reasoning_content %}
1061
+ {%- else %}
1062
+ {%- if '</think>' in content %}
1063
+ {%- set reasoning_content = content.split('</think>')[0].strip('\n').split('<think>')[-1].strip('\n') %}
1064
+ {%- set content = content.split('</think>')[-1].strip('\n') %}
1065
+ {%- endif %}
1066
+ {%- endif %}
1067
+ {%- if reasoning_content and loop.index0 > ns.last_user_index -%}
1068
+ {{- '<think>' ~ '\n' ~ reasoning_content ~ '\n' ~ '</think>' ~ '\n\n' }}
1069
+ {%- endif -%}
1070
+ {%- if content -%}
1071
+ {{- content }}
1072
+ {%- endif -%}
1073
+ {%- if message.tool_calls -%}
1074
+ {{- '\n' ~ toolcall_begin_token ~ '\n' }}
1075
+
1076
+ {%- for tool_call in message.tool_calls -%}
1077
+ {%- if tool_call.function %}
1078
+ {%- set tool_call = tool_call.function %}
1079
+ {%- endif %}
1080
+ {{- '<invoke name="' + tool_call.name + '">' }}
1081
+ {% set _args = tool_call.arguments %}
1082
+ {%- for k, v in _args.items() %}
1083
+ {{- '<parameter name="' + k + '">' }}
1084
+ {{- v | tojson(ensure_ascii=False) if v is not string else v }}
1085
+ {{- '</parameter>' }}
1086
+ {% endfor %}
1087
+ {{- '</invoke>' ~ '\n' }}
1088
+ {%- endfor -%}
1089
+
1090
+ {{- toolcall_end_token}}
1091
+ {%- set last_tool_call.name = message.tool_calls[-1].name -%}
1092
+ {%- else -%}
1093
+ {%- set last_tool_call.name = none -%}
1094
+ {%- endif -%}
1095
+ {{- '[e~[' ~ '\n' }}
1096
+
1097
+ {%- elif message.role == 'tool' -%}
1098
+ {%- if last_tool_call.name is none -%}
1099
+ {{- raise_exception("Message has tool role, but there was no previous assistant message with a tool call!") }}
1100
+ {%- endif -%}
1101
+ {%- if loop.first or (conversation_messages[loop.index0 - 1].role != 'tool') -%}
1102
+ {{- ']~b]tool' }}
1103
+ {%- endif -%}
1104
+ {%- if message.content is string -%}
1105
+ {{- '\n<response>' }}
1106
+ {{- message.content }}
1107
+ {{- '</response>' }}
1108
+ {%- else -%}
1109
+ {%- for tr in message.content -%}
1110
+ {{- '\n<response>' }}
1111
+ {{- tr.output if tr.output is defined else (tr.text if tr.type == 'text' and tr.text is defined else tr) }}
1112
+ {{- '\n</response>' }}
1113
+ {%- endfor -%}
1114
+ {%- endif -%}
1115
+ {%- if loop.last or (conversation_messages[loop.index0 + 1].role != 'tool') -%}
1116
+ {{- '[e~[\n' -}}
1117
+ {%- endif -%}
1118
+
1119
+ {%- elif message.role == 'user' -%}
1120
+ {{- ']~b]user' ~ '\n' }}
1121
+ {{- visible_text(message.content) }}
1122
+ {{- '[e~[' ~ '\n' }}
1123
+ {%- endif -%}
1124
+ {%- endfor -%}
1125
+
1126
+ {#- Generation prompt -#}
1127
+ {%- if add_generation_prompt -%}
1128
+ {{- ']~b]ai' ~ '\n' ~ '<think>' ~ '\n' }}
1129
+ {%- endif -%}
1130
+
1131
+ INFO:gguf.gguf_writer:Writing the following files:
1132
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00001-of-00010.gguf: n_tensors = 90, total_size = 47.8G
1133
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00002-of-00010.gguf: n_tensors = 90, total_size = 49.0G
1134
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00003-of-00010.gguf: n_tensors = 80, total_size = 48.9G
1135
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00004-of-00010.gguf: n_tensors = 90, total_size = 49.0G
1136
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00005-of-00010.gguf: n_tensors = 90, total_size = 49.0G
1137
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00006-of-00010.gguf: n_tensors = 80, total_size = 48.9G
1138
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00007-of-00010.gguf: n_tensors = 90, total_size = 49.0G
1139
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00008-of-00010.gguf: n_tensors = 90, total_size = 49.0G
1140
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00009-of-00010.gguf: n_tensors = 80, total_size = 48.9G
1141
+ INFO:gguf.gguf_writer:/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00010-of-00010.gguf: n_tensors = 29, total_size = 18.3G
1142
+
1143
+
1144
+
1145
+
1146
+
1147
+ INFO:hf-to-gguf:Model successfully exported to /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/
logs/imatrix-MiniMax-M2.7-BF16.log ADDED
@@ -0,0 +1,819 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model=/mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00001-of-00010.gguf
2
+
3
+ numactl -N ${SOCKET} -m ${SOCKET} \
4
+ ./build/bin/llama-imatrix \
5
+ --model "$model"\
6
+ -f ubergarm-imatrix-calibration-corpus-v02.txt \
7
+ -o /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat \
8
+ --no-fused-moe \
9
+ --no-fused-up-gate \
10
+ --no-fused-mul-multiadd \
11
+ --ctx-size 512 \
12
+ -ub 4096 -b 4096 \
13
+ --threads 96 \
14
+ --threads-batch 128 \
15
+ --no-mmap \
16
+ --numa numactl \
17
+ --verbosity 1 \
18
+ --layer-similarity
19
+
20
+ CPU: using device CPU - 0 MiB free
21
+ llama_model_loader: additional 9 GGUFs metadata loaded.
22
+ llama_model_loader: loaded meta data with 40 key-value pairs and 809 tensors from /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/MiniMax-M2.7-256x4.9B-BF16-00001-of-00010.gguf (version GGUF V3 (latest))
23
+ llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
24
+ llama_model_loader: - kv 0: general.architecture str = minimax-m2
25
+ llama_model_loader: - kv 1: general.type str = model
26
+ llama_model_loader: - kv 2: general.sampling.top_k i32 = 40
27
+ llama_model_loader: - kv 3: general.sampling.top_p f32 = 0.950000
28
+ llama_model_loader: - kv 4: general.sampling.temp f32 = 1.000000
29
+ llama_model_loader: - kv 5: general.name str = MiniMax M2.7
30
+ llama_model_loader: - kv 6: general.size_label str = 256x4.9B
31
+ llama_model_loader: - kv 7: general.license str = other
32
+ llama_model_loader: - kv 8: general.license.name str = modified-mit
33
+ llama_model_loader: - kv 9: general.license.link str = https://github.com/MiniMax-AI/MiniMax...
34
+ llama_model_loader: - kv 10: general.tags arr[str,1] = ["text-generation"]
35
+ llama_model_loader: - kv 11: minimax-m2.block_count u32 = 62
36
+ llama_model_loader: - kv 12: minimax-m2.context_length u32 = 196608
37
+ llama_model_loader: - kv 13: minimax-m2.embedding_length u32 = 3072
38
+ llama_model_loader: - kv 14: minimax-m2.feed_forward_length u32 = 1536
39
+ llama_model_loader: - kv 15: minimax-m2.attention.head_count u32 = 48
40
+ llama_model_loader: - kv 16: minimax-m2.attention.head_count_kv u32 = 8
41
+ llama_model_loader: - kv 17: minimax-m2.rope.freq_base f32 = 5000000.000000
42
+ llama_model_loader: - kv 18: minimax-m2.attention.layer_norm_rms_epsilon f32 = 0.000001
43
+ llama_model_loader: - kv 19: minimax-m2.expert_count u32 = 256
44
+ llama_model_loader: - kv 20: minimax-m2.expert_used_count u32 = 8
45
+ llama_model_loader: - kv 21: minimax-m2.expert_gating_func u32 = 2
46
+ llama_model_loader: - kv 22: minimax-m2.attention.key_length u32 = 128
47
+ llama_model_loader: - kv 23: minimax-m2.attention.value_length u32 = 128
48
+ llama_model_loader: - kv 24: general.file_type u32 = 32
49
+ llama_model_loader: - kv 25: minimax-m2.expert_feed_forward_length u32 = 1536
50
+ llama_model_loader: - kv 26: minimax-m2.rope.dimension_count u32 = 64
51
+ llama_model_loader: - kv 27: general.quantization_version u32 = 2
52
+ llama_model_loader: - kv 28: tokenizer.ggml.model str = gpt2
53
+ llama_model_loader: - kv 29: tokenizer.ggml.pre str = minimax-m2
54
+ llama_model_loader: - kv 30: tokenizer.ggml.tokens arr[str,200064] = ["Ā", "ā", "Ă", "ă", "Ą", "ą", ...
55
+ llama_model_loader: - kv 31: tokenizer.ggml.token_type arr[i32,200064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
56
+ llama_model_loader: - kv 32: tokenizer.ggml.merges arr[str,199744] = ["Ġ Ġ", "Ġ t", "Ġ a", "i n", "e r...
57
+ llama_model_loader: - kv 33: tokenizer.ggml.bos_token_id u32 = 200034
58
+ llama_model_loader: - kv 34: tokenizer.ggml.eos_token_id u32 = 200020
59
+ llama_model_loader: - kv 35: tokenizer.ggml.unknown_token_id u32 = 200021
60
+ llama_model_loader: - kv 36: tokenizer.chat_template str = {# ----------‑‑‑ special token ...
61
+ llama_model_loader: - kv 37: split.no u16 = 0
62
+ llama_model_loader: - kv 38: split.count u16 = 10
63
+ llama_model_loader: - kv 39: split.tensors.count i32 = 809
64
+ llama_model_loader: - type f32: 373 tensors
65
+ llama_model_loader: - type bf16: 436 tensors
66
+ load: 0 unused tokens
67
+ load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
68
+ load: printing all EOG tokens:
69
+ load: - 200004 ('<fim_pad>')
70
+ load: - 200005 ('<reponame>')
71
+ load: - 200020 ('[e~[')
72
+ load: special tokens cache size = 54
73
+ load: token to piece cache size = 1.3355 MB
74
+ llm_load_print_meta: format = GGUF V3 (latest)
75
+ llm_load_print_meta: arch = minimax-m2
76
+ llm_load_print_meta: n_ctx_train = 196608
77
+ llm_load_print_meta: n_embd = 3072
78
+ llm_load_print_meta: n_layer = 62
79
+ llm_load_print_meta: n_head = 48
80
+ llm_load_print_meta: n_head_kv = 8
81
+ llm_load_print_meta: n_rot = 64
82
+ llm_load_print_meta: n_swa = 0
83
+ llm_load_print_meta: n_swa_pattern = 1
84
+ llm_load_print_meta: n_embd_head_k = 128
85
+ llm_load_print_meta: n_embd_head_v = 128
86
+ llm_load_print_meta: n_gqa = 6
87
+ llm_load_print_meta: n_embd_k_gqa = 1024
88
+ llm_load_print_meta: n_embd_v_gqa = 1024
89
+ llm_load_print_meta: f_norm_eps = 0.0e+00
90
+ llm_load_print_meta: f_norm_rms_eps = 1.0e-06
91
+ llm_load_print_meta: f_clamp_kqv = 0.0e+00
92
+ llm_load_print_meta: f_max_alibi_bias = 0.0e+00
93
+ llm_load_print_meta: f_logit_scale = 0.0e+00
94
+ llm_load_print_meta: n_ff = 1536
95
+ llm_load_print_meta: n_expert = 256
96
+ llm_load_print_meta: n_expert_used = 8
97
+ llm_load_print_meta: causal attn = 1
98
+ llm_load_print_meta: pooling type = 0
99
+ llm_load_print_meta: rope type = 2
100
+ llm_load_print_meta: rope scaling = linear
101
+ llm_load_print_meta: freq_base_train = 5000000.0
102
+ llm_load_print_meta: freq_scale_train = 1
103
+ llm_load_print_meta: n_ctx_orig_yarn = 196608
104
+ llm_load_print_meta: rope_finetuned = unknown
105
+ llm_load_print_meta: ssm_d_conv = 0
106
+ llm_load_print_meta: ssm_d_inner = 0
107
+ llm_load_print_meta: ssm_d_state = 0
108
+ llm_load_print_meta: ssm_dt_rank = 0
109
+ llm_load_print_meta: ssm_n_group = 0
110
+ llm_load_print_meta: model type = 230B.A10B
111
+ llm_load_print_meta: model ftype = BF16
112
+ llm_load_print_meta: model params = 228.690 B
113
+ llm_load_print_meta: model size = 426.060 GiB (16.003 BPW)
114
+ llm_load_print_meta: repeating layers = 423.771 GiB (16.003 BPW, 227.461 B parameters)
115
+ llm_load_print_meta: general.name = MiniMax M2.7
116
+ print_info: vocab type = BPE
117
+ print_info: n_vocab = 200064
118
+ print_info: n_merges = 199744
119
+ print_info: BOS token = 200034 ']~!b['
120
+ print_info: EOS token = 200020 '[e~['
121
+ print_info: UNK token = 200021 ']!d~['
122
+ print_info: LF token = 10 'Ċ'
123
+ print_info: FIM PRE token = 200001 '<fim_prefix>'
124
+ print_info: FIM SUF token = 200003 '<fim_suffix>'
125
+ print_info: FIM MID token = 200002 '<fim_middle>'
126
+ print_info: FIM PAD token = 200004 '<fim_pad>'
127
+ print_info: FIM REP token = 200005 '<reponame>'
128
+ print_info: EOG token = 200004 '<fim_pad>'
129
+ print_info: EOG token = 200005 '<reponame>'
130
+ print_info: EOG token = 200020 '[e~['
131
+ print_info: max token length = 256
132
+ ======================================= HAVE_FANCY_SIMD is defined
133
+ Free memory 0 MiB on device 0 is less the 1024 MiB safety margin
134
+ ------------------- Layer sizes:
135
+ Layer 0: 6999.05, 2.00, 7001.05 108.00 MiB
136
+ Layer 1: 6999.05, 2.00, 7001.05 108.00 MiB
137
+ Layer 2: 6999.05, 2.00, 7001.05 108.00 MiB
138
+ Layer 3: 6999.05, 2.00, 7001.05 108.00 MiB
139
+ Layer 4: 6999.05, 2.00, 7001.05 108.00 MiB
140
+ Layer 5: 6999.05, 2.00, 7001.05 108.00 MiB
141
+ Layer 6: 6999.05, 2.00, 7001.05 108.00 MiB
142
+ Layer 7: 6999.05, 2.00, 7001.05 108.00 MiB
143
+ Layer 8: 6999.05, 2.00, 7001.05 108.00 MiB
144
+ Layer 9: 6999.05, 2.00, 7001.05 108.00 MiB
145
+ Layer 10: 6999.05, 2.00, 7001.05 108.00 MiB
146
+ Layer 11: 6999.05, 2.00, 7001.05 108.00 MiB
147
+ Layer 12: 6999.05, 2.00, 7001.05 108.00 MiB
148
+ Layer 13: 6999.05, 2.00, 7001.05 108.00 MiB
149
+ Layer 14: 6999.05, 2.00, 7001.05 108.00 MiB
150
+ Layer 15: 6999.05, 2.00, 7001.05 108.00 MiB
151
+ Layer 16: 6999.05, 2.00, 7001.05 108.00 MiB
152
+ Layer 17: 6999.05, 2.00, 7001.05 108.00 MiB
153
+ Layer 18: 6999.05, 2.00, 7001.05 108.00 MiB
154
+ Layer 19: 6999.05, 2.00, 7001.05 108.00 MiB
155
+ Layer 20: 6999.05, 2.00, 7001.05 108.00 MiB
156
+ Layer 21: 6999.05, 2.00, 7001.05 108.00 MiB
157
+ Layer 22: 6999.05, 2.00, 7001.05 108.00 MiB
158
+ Layer 23: 6999.05, 2.00, 7001.05 108.00 MiB
159
+ Layer 24: 6999.05, 2.00, 7001.05 108.00 MiB
160
+ Layer 25: 6999.05, 2.00, 7001.05 108.00 MiB
161
+ Layer 26: 6999.05, 2.00, 7001.05 108.00 MiB
162
+ Layer 27: 6999.05, 2.00, 7001.05 108.00 MiB
163
+ Layer 28: 6999.05, 2.00, 7001.05 108.00 MiB
164
+ Layer 29: 6999.05, 2.00, 7001.05 108.00 MiB
165
+ Layer 30: 6999.05, 2.00, 7001.05 108.00 MiB
166
+ Layer 31: 6999.05, 2.00, 7001.05 108.00 MiB
167
+ Layer 32: 6999.05, 2.00, 7001.05 108.00 MiB
168
+ Layer 33: 6999.05, 2.00, 7001.05 108.00 MiB
169
+ Layer 34: 6999.05, 2.00, 7001.05 108.00 MiB
170
+ Layer 35: 6999.05, 2.00, 7001.05 108.00 MiB
171
+ Layer 36: 6999.05, 2.00, 7001.05 108.00 MiB
172
+ Layer 37: 6999.05, 2.00, 7001.05 108.00 MiB
173
+ Layer 38: 6999.05, 2.00, 7001.05 108.00 MiB
174
+ Layer 39: 6999.05, 2.00, 7001.05 108.00 MiB
175
+ Layer 40: 6999.05, 2.00, 7001.05 108.00 MiB
176
+ Layer 41: 6999.05, 2.00, 7001.05 108.00 MiB
177
+ Layer 42: 6999.05, 2.00, 7001.05 108.00 MiB
178
+ Layer 43: 6999.05, 2.00, 7001.05 108.00 MiB
179
+ Layer 44: 6999.05, 2.00, 7001.05 108.00 MiB
180
+ Layer 45: 6999.05, 2.00, 7001.05 108.00 MiB
181
+ Layer 46: 6999.05, 2.00, 7001.05 108.00 MiB
182
+ Layer 47: 6999.05, 2.00, 7001.05 108.00 MiB
183
+ Layer 48: 6999.05, 2.00, 7001.05 108.00 MiB
184
+ Layer 49: 6999.05, 2.00, 7001.05 108.00 MiB
185
+ Layer 50: 6999.05, 2.00, 7001.05 108.00 MiB
186
+ Layer 51: 6999.05, 2.00, 7001.05 108.00 MiB
187
+ Layer 52: 6999.05, 2.00, 7001.05 108.00 MiB
188
+ Layer 53: 6999.05, 2.00, 7001.05 108.00 MiB
189
+ Layer 54: 6999.05, 2.00, 7001.05 108.00 MiB
190
+ Layer 55: 6999.05, 2.00, 7001.05 108.00 MiB
191
+ Layer 56: 6999.05, 2.00, 7001.05 108.00 MiB
192
+ Layer 57: 6999.05, 2.00, 7001.05 108.00 MiB
193
+ Layer 58: 6999.05, 2.00, 7001.05 108.00 MiB
194
+ Layer 59: 6999.05, 2.00, 7001.05 108.00 MiB
195
+ Layer 60: 6999.05, 2.00, 7001.05 108.00 MiB
196
+ Layer 61: 6999.05, 2.00, 7001.05 108.00 MiB
197
+ Layer 62: 1172.25, 282.75, 1455.00 MiB (output layer)
198
+ --------------------------------------------------------------------------
199
+ Total : 433941.21, 406.75, 434347.96 MiB
200
+ Free memory 0 MiB on device 0 is less the required compute buffer size 108 MiB
201
+ Memory required for model tensors + cache: 435520 MiB
202
+ Memory available on all devices - compute: 0 MiB
203
+ llm_load_tensors: ggml ctx size = 0.35 MiB
204
+ llm_load_tensors: offloading 0 repeating layers to GPU
205
+ llm_load_tensors: offloaded 0/63 layers to GPU
206
+ llm_load_tensors: CPU buffer size = 436285.72 MiB
207
+ ....................................................................................................
208
+ llama_init_from_model: n_ctx = 512
209
+ llama_init_from_model: n_batch = 512
210
+ llama_init_from_model: n_ubatch = 512
211
+ llama_init_from_model: flash_attn = 1
212
+ llama_init_from_model: attn_max_b = 0
213
+ llama_init_from_model: fused_moe = 0
214
+ llama_init_from_model: grouped er = 0
215
+ llama_init_from_model: fused_up_gate = 0
216
+ llama_init_from_model: fused_mmad = 0
217
+ llama_init_from_model: rope_cache = 0
218
+ llama_init_from_model: graph_reuse = 1
219
+ llama_init_from_model: k_cache_hadam = 0
220
+ llama_init_from_model: v_cache_hadam = 0
221
+ llama_init_from_model: split_mode_graph_scheduling = 0
222
+ llama_init_from_model: reduce_type = f16
223
+ llama_init_from_model: sched_async = 0
224
+ llama_init_from_model: ser = -1, 0
225
+ llama_init_from_model: freq_base = 5000000.0
226
+ llama_init_from_model: freq_scale = 1
227
+ llama_kv_cache_init: CPU KV buffer size = 124.00 MiB
228
+ llama_init_from_model: KV self size = 124.00 MiB, K (f16): 62.00 MiB, V (f16): 62.00 MiB
229
+ llama_init_from_model: CPU output buffer size = 0.76 MiB
230
+ llama_init_from_model: CPU compute buffer size = 402.75 MiB
231
+ llama_init_from_model: graph nodes = 2609
232
+ llama_init_from_model: graph splits = 1
233
+ llama_init_from_model: enabling only_active_experts scheduling
234
+
235
+ system_info: n_threads = 96 (n_threads_batch = 128) / 512 | AVX = 1 | AVX_VNNI = 1 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | AVX512_BF16 = 1 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 |
236
+ compute_imatrix: tokenizing the input ..
237
+ compute_imatrix: tokenization took 779.628 ms
238
+ compute_imatrix: computing over 796 chunks with batch_size 512
239
+ compute_imatrix: 3.99 seconds per pass - ETA 52.95 minutes
240
+ [1]91.6530,[2]16.4837,[3]7.9862,[4]4.7587,[5]3.6240,[6]2.9956,[7]2.5865,[8]2.3337,[9]2.2805,
241
+ save_imatrix: entry ' blk.60.ffn_down_exps.weight' has partial data (91.80%) 21 out of 256 experts are missing data - skipping
242
+ save_imatrix: entry ' blk.59.ffn_gate_exps.weight' has partial data (94.14%) 15 out of 256 experts are missing data - skipping
243
+ save_imatrix: entry ' blk.59.ffn_up_exps.weight' has partial data (94.14%) 15 out of 256 experts are missing data - skipping
244
+ save_imatrix: entry ' blk.58.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
245
+ save_imatrix: entry ' blk.58.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
246
+ save_imatrix: entry ' blk.57.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
247
+ save_imatrix: entry ' blk.57.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
248
+ save_imatrix: entry ' blk.57.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
249
+ save_imatrix: entry ' blk.56.ffn_gate_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
250
+ save_imatrix: entry ' blk.55.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
251
+ save_imatrix: entry ' blk.60.ffn_gate_exps.weight' has partial data (91.80%) 21 out of 256 experts are missing data - skipping
252
+ save_imatrix: entry ' blk.55.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
253
+ save_imatrix: entry ' blk.55.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
254
+ save_imatrix: entry ' blk.54.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
255
+ save_imatrix: entry ' blk.53.ffn_down_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
256
+ save_imatrix: entry ' blk.56.ffn_down_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
257
+ save_imatrix: entry ' blk.52.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
258
+ save_imatrix: entry ' blk.52.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
259
+ save_imatrix: entry ' blk.52.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
260
+ save_imatrix: entry ' blk.48.ffn_up_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
261
+ save_imatrix: entry ' blk.47.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
262
+ save_imatrix: entry ' blk.47.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
263
+ save_imatrix: entry ' blk.46.ffn_down_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
264
+ save_imatrix: entry ' blk.46.ffn_up_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
265
+ save_imatrix: entry ' blk.45.ffn_down_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
266
+ save_imatrix: entry ' blk.43.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
267
+ save_imatrix: entry ' blk.43.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
268
+ save_imatrix: entry ' blk.42.ffn_down_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
269
+ save_imatrix: entry ' blk.42.ffn_up_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
270
+ save_imatrix: entry ' blk.44.ffn_gate_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
271
+ save_imatrix: entry ' blk.39.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
272
+ save_imatrix: entry ' blk.38.ffn_down_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
273
+ save_imatrix: entry ' blk.38.ffn_gate_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
274
+ save_imatrix: entry ' blk.39.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
275
+ save_imatrix: entry ' blk.37.ffn_gate_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
276
+ save_imatrix: entry ' blk.37.ffn_up_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
277
+ save_imatrix: entry ' blk.36.ffn_down_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
278
+ save_imatrix: entry ' blk.40.ffn_down_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
279
+ save_imatrix: entry ' blk.35.ffn_down_exps.weight' has partial data (93.75%) 16 out of 256 experts are missing data - skipping
280
+ save_imatrix: entry ' blk.35.ffn_gate_exps.weight' has partial data (93.75%) 16 out of 256 experts are missing data - skipping
281
+ save_imatrix: entry ' blk.51.ffn_gate_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
282
+ save_imatrix: entry ' blk.34.ffn_up_exps.weight' has partial data (92.58%) 19 out of 256 experts are missing data - skipping
283
+ save_imatrix: entry ' blk.33.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
284
+ save_imatrix: entry ' blk.33.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
285
+ save_imatrix: entry ' blk.33.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
286
+ save_imatrix: entry ' blk.39.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
287
+ save_imatrix: entry ' blk.32.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
288
+ save_imatrix: entry ' blk.32.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
289
+ save_imatrix: entry ' blk.34.ffn_down_exps.weight' has partial data (92.58%) 19 out of 256 experts are missing data - skipping
290
+ save_imatrix: entry ' blk.42.ffn_gate_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
291
+ save_imatrix: entry ' blk.50.ffn_down_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
292
+ save_imatrix: entry ' blk.15.ffn_gate_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
293
+ save_imatrix: entry ' blk.38.ffn_up_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
294
+ save_imatrix: entry ' blk.45.ffn_gate_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
295
+ save_imatrix: entry ' blk.16.ffn_up_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
296
+ save_imatrix: entry ' blk.47.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
297
+ save_imatrix: entry ' blk.45.ffn_up_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
298
+ save_imatrix: entry ' blk.23.ffn_gate_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
299
+ save_imatrix: entry ' blk.15.ffn_down_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
300
+ save_imatrix: entry ' blk.16.ffn_gate_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
301
+ save_imatrix: entry ' blk.51.ffn_up_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
302
+ save_imatrix: entry ' blk.53.ffn_gate_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
303
+ save_imatrix: entry ' blk.41.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
304
+ save_imatrix: entry ' blk.50.ffn_up_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
305
+ save_imatrix: entry ' blk.59.ffn_down_exps.weight' has partial data (94.14%) 15 out of 256 experts are missing data - skipping
306
+ save_imatrix: entry ' blk.54.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
307
+ save_imatrix: entry ' blk.4.ffn_up_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
308
+ save_imatrix: entry ' blk.44.ffn_down_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
309
+ save_imatrix: entry ' blk.36.ffn_gate_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
310
+ save_imatrix: entry ' blk.31.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
311
+ save_imatrix: entry ' blk.58.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
312
+ save_imatrix: entry ' blk.29.ffn_gate_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
313
+ save_imatrix: entry ' blk.41.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
314
+ save_imatrix: entry ' blk.6.ffn_up_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
315
+ save_imatrix: entry ' blk.41.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
316
+ save_imatrix: entry ' blk.1.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
317
+ save_imatrix: entry ' blk.34.ffn_gate_exps.weight' has partial data (92.58%) 19 out of 256 experts are missing data - skipping
318
+ save_imatrix: entry ' blk.49.ffn_down_exps.weight' has partial data (98.05%) 5 out of 256 experts are missing data Storing **but be aware**
319
+ save_imatrix: entry ' blk.19.ffn_up_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
320
+ save_imatrix: entry ' blk.15.ffn_up_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
321
+ save_imatrix: entry ' blk.56.ffn_up_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
322
+ save_imatrix: entry ' blk.44.ffn_up_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
323
+ save_imatrix: entry ' blk.4.ffn_gate_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
324
+ save_imatrix: entry ' blk.23.ffn_down_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
325
+ save_imatrix: entry ' blk.0.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
326
+ save_imatrix: entry ' blk.37.ffn_down_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
327
+ save_imatrix: entry ' blk.32.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
328
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (81.64%) 47 out of 256 experts are missing data - skipping
329
+ save_imatrix: entry ' blk.48.ffn_down_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
330
+ save_imatrix: entry ' blk.29.ffn_up_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
331
+ save_imatrix: entry ' blk.0.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
332
+ save_imatrix: entry ' blk.40.ffn_up_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
333
+ save_imatrix: entry ' blk.18.ffn_up_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
334
+ save_imatrix: entry ' blk.21.ffn_down_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
335
+ save_imatrix: entry ' blk.4.ffn_down_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
336
+ save_imatrix: entry ' blk.54.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
337
+ save_imatrix: entry ' blk.2.ffn_up_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
338
+ save_imatrix: entry ' blk.28.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
339
+ save_imatrix: entry ' blk.50.ffn_gate_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
340
+ save_imatrix: entry ' blk.24.ffn_down_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
341
+ save_imatrix: entry ' blk.28.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
342
+ save_imatrix: entry ' blk.51.ffn_down_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
343
+ save_imatrix: entry ' blk.6.ffn_down_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
344
+ save_imatrix: entry ' blk.26.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
345
+ save_imatrix: entry ' blk.30.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
346
+ save_imatrix: entry ' blk.40.ffn_gate_exps.weight' has partial data (94.53%) 14 out of 256 experts are missing data - skipping
347
+ save_imatrix: entry ' blk.1.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
348
+ save_imatrix: entry ' blk.1.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
349
+ save_imatrix: entry ' blk.16.ffn_down_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
350
+ save_imatrix: entry ' blk.18.ffn_gate_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
351
+ save_imatrix: entry ' blk.49.ffn_up_exps.weight' has partial data (98.05%) 5 out of 256 experts are missing data Storing **but be aware**
352
+ save_imatrix: entry ' blk.35.ffn_up_exps.weight' has partial data (93.75%) 16 out of 256 experts are missing data - skipping
353
+ save_imatrix: entry ' blk.18.ffn_down_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
354
+ save_imatrix: entry ' blk.21.ffn_up_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
355
+ save_imatrix: entry ' blk.6.ffn_gate_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
356
+ save_imatrix: entry ' blk.19.ffn_gate_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
357
+ save_imatrix: entry ' blk.24.ffn_up_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
358
+ save_imatrix: entry ' blk.19.ffn_down_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
359
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (81.64%) 47 out of 256 experts are missing data - skipping
360
+ save_imatrix: entry ' blk.2.ffn_down_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
361
+ save_imatrix: entry ' blk.48.ffn_gate_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
362
+ save_imatrix: entry ' blk.21.ffn_gate_exps.weight' has partial data (99.22%) 2 out of 256 experts are missing data Storing **but be aware**
363
+ save_imatrix: entry ' blk.29.ffn_down_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
364
+ save_imatrix: entry ' blk.22.ffn_up_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
365
+ save_imatrix: entry ' blk.22.ffn_gate_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
366
+ save_imatrix: entry ' blk.22.ffn_down_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
367
+ save_imatrix: entry ' blk.23.ffn_up_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
368
+ save_imatrix: entry ' blk.53.ffn_up_exps.weight' has partial data (98.83%) 3 out of 256 experts are missing data Storing **but be aware**
369
+ save_imatrix: entry ' blk.24.ffn_gate_exps.weight' has partial data (97.27%) 7 out of 256 experts are missing data Storing **but be aware**
370
+ save_imatrix: entry ' blk.49.ffn_gate_exps.weight' has partial data (98.05%) 5 out of 256 experts are missing data Storing **but be aware**
371
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (81.64%) 47 out of 256 experts are missing data - skipping
372
+ save_imatrix: entry ' blk.25.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
373
+ save_imatrix: entry ' blk.2.ffn_gate_exps.weight' has partial data (99.61%) 1 out of 256 experts are missing data Storing **but be aware**
374
+ save_imatrix: entry ' blk.25.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
375
+ save_imatrix: entry ' blk.25.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
376
+ save_imatrix: entry ' blk.36.ffn_up_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
377
+ save_imatrix: entry ' blk.26.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
378
+ save_imatrix: entry ' blk.27.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
379
+ save_imatrix: entry ' blk.27.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
380
+ save_imatrix: entry ' blk.60.ffn_up_exps.weight' has partial data (91.80%) 21 out of 256 experts are missing data - skipping
381
+ save_imatrix: entry ' blk.26.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
382
+ save_imatrix: entry ' blk.27.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
383
+ save_imatrix: entry ' blk.43.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
384
+ save_imatrix: entry ' blk.0.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
385
+ save_imatrix: entry ' blk.28.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
386
+ save_imatrix: entry ' blk.31.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
387
+ save_imatrix: entry ' blk.46.ffn_gate_exps.weight' has partial data (98.44%) 4 out of 256 experts are missing data Storing **but be aware**
388
+ save_imatrix: entry ' blk.30.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
389
+ save_imatrix: entry ' blk.30.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
390
+ save_imatrix: entry ' blk.31.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
391
+ save_imatrix: warning: storing only 461 out of 497 entries
392
+
393
+ save_imatrix: stored collected data after 10 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
394
+ [10]2.2152,[11]2.2249,[12]2.4598,[13]2.5706,[14]2.5471,[15]2.4061,[16]2.2934,[17]2.1901,[18]2.1276,[19]2.0519,
395
+ save_imatrix: entry ' blk.60.ffn_down_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
396
+ save_imatrix: entry ' blk.59.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
397
+ save_imatrix: entry ' blk.59.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
398
+ save_imatrix: entry ' blk.60.ffn_gate_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
399
+ save_imatrix: entry ' blk.45.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
400
+ save_imatrix: entry ' blk.44.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
401
+ save_imatrix: entry ' blk.38.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
402
+ save_imatrix: entry ' blk.38.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
403
+ save_imatrix: entry ' blk.37.ffn_gate_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
404
+ save_imatrix: entry ' blk.37.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
405
+ save_imatrix: entry ' blk.36.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
406
+ save_imatrix: entry ' blk.40.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
407
+ save_imatrix: entry ' blk.35.ffn_down_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
408
+ save_imatrix: entry ' blk.35.ffn_gate_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
409
+ save_imatrix: entry ' blk.34.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
410
+ save_imatrix: entry ' blk.34.ffn_down_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
411
+ save_imatrix: entry ' blk.38.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
412
+ save_imatrix: entry ' blk.45.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
413
+ save_imatrix: entry ' blk.45.ffn_up_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
414
+ save_imatrix: entry ' blk.59.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
415
+ save_imatrix: entry ' blk.44.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
416
+ save_imatrix: entry ' blk.36.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
417
+ save_imatrix: entry ' blk.29.ffn_gate_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
418
+ save_imatrix: entry ' blk.34.ffn_gate_exps.weight' has partial data (95.31%) 12 out of 256 experts are missing data Storing **but be aware**
419
+ save_imatrix: entry ' blk.44.ffn_up_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
420
+ save_imatrix: entry ' blk.37.ffn_down_exps.weight' has partial data (96.09%) 10 out of 256 experts are missing data Storing **but be aware**
421
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (83.59%) 42 out of 256 experts are missing data - skipping
422
+ save_imatrix: entry ' blk.29.ffn_up_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
423
+ save_imatrix: entry ' blk.40.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
424
+ save_imatrix: entry ' blk.40.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
425
+ save_imatrix: entry ' blk.35.ffn_up_exps.weight' has partial data (95.70%) 11 out of 256 experts are missing data Storing **but be aware**
426
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (83.59%) 42 out of 256 experts are missing data - skipping
427
+ save_imatrix: entry ' blk.29.ffn_down_exps.weight' has partial data (97.66%) 6 out of 256 experts are missing data Storing **but be aware**
428
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (83.59%) 42 out of 256 experts are missing data - skipping
429
+ save_imatrix: entry ' blk.36.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
430
+ save_imatrix: entry ' blk.60.ffn_up_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
431
+ save_imatrix: warning: storing only 491 out of 497 entries
432
+
433
+ save_imatrix: stored collected data after 20 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
434
+ [20]2.0097,[21]1.9543,[22]1.9059,[23]1.9143,[24]1.8992,[25]1.8525,[26]1.9506,[27]2.0486,[28]2.1579,[29]2.1379,
435
+ save_imatrix: entry ' blk.60.ffn_down_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
436
+ save_imatrix: entry ' blk.60.ffn_gate_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
437
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (83.98%) 41 out of 256 experts are missing data - skipping
438
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (83.98%) 41 out of 256 experts are missing data - skipping
439
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (83.98%) 41 out of 256 experts are missing data - skipping
440
+ save_imatrix: entry ' blk.60.ffn_up_exps.weight' has partial data (94.92%) 13 out of 256 experts are missing data - skipping
441
+ save_imatrix: warning: storing only 491 out of 497 entries
442
+
443
+ save_imatrix: stored collected data after 30 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
444
+ [30]2.3057,[31]2.2805,[32]2.3091,[33]2.2842,[34]2.2976,[35]2.3024,[36]2.2904,[37]2.2937,[38]2.3360,[39]2.3423,
445
+ save_imatrix: entry ' blk.60.ffn_down_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
446
+ save_imatrix: entry ' blk.60.ffn_gate_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
447
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (87.11%) 33 out of 256 experts are missing data - skipping
448
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (87.11%) 33 out of 256 experts are missing data - skipping
449
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (87.11%) 33 out of 256 experts are missing data - skipping
450
+ save_imatrix: entry ' blk.60.ffn_up_exps.weight' has partial data (96.48%) 9 out of 256 experts are missing data Storing **but be aware**
451
+ save_imatrix: warning: storing only 494 out of 497 entries
452
+
453
+ save_imatrix: stored collected data after 40 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
454
+ [40]2.3262,[41]2.3632,[42]2.3749,[43]2.3947,[44]2.4239,[45]2.4425,[46]2.4257,[47]2.4243,[48]2.4192,[49]2.4269,
455
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (87.50%) 32 out of 256 experts are missing data - skipping
456
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (87.50%) 32 out of 256 experts are missing data - skipping
457
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (87.50%) 32 out of 256 experts are missing data - skipping
458
+ save_imatrix: warning: storing only 494 out of 497 entries
459
+
460
+ save_imatrix: stored collected data after 50 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
461
+ [50]2.4123,[51]2.4286,[52]2.4539,[53]2.4376,[54]2.4462,[55]2.4430,[56]2.4412,[57]2.4322,[58]2.5168,[59]2.5695,
462
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (89.06%) 28 out of 256 experts are missing data - skipping
463
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (89.06%) 28 out of 256 experts are missing data - skipping
464
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (89.06%) 28 out of 256 experts are missing data - skipping
465
+ save_imatrix: warning: storing only 494 out of 497 entries
466
+
467
+ save_imatrix: stored collected data after 60 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
468
+ [60]2.6236,[61]2.6529,[62]2.7734,[63]2.8045,[64]2.8668,[65]2.9429,[66]3.0196,[67]3.1169,[68]3.2083,[69]3.2922,
469
+ save_imatrix: entry ' blk.61.ffn_down_exps.weight' has partial data (96.88%) 8 out of 256 experts are missing data Storing **but be aware**
470
+ save_imatrix: entry ' blk.61.ffn_up_exps.weight' has partial data (96.88%) 8 out of 256 experts are missing data Storing **but be aware**
471
+ save_imatrix: entry ' blk.61.ffn_gate_exps.weight' has partial data (96.88%) 8 out of 256 experts are missing data Storing **but be aware**
472
+
473
+ save_imatrix: stored collected data after 70 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
474
+ [70]3.3446,[71]3.3814,[72]3.3968,[73]3.4427,[74]3.5064,[75]3.6009,[76]3.5929,[77]3.5709,[78]3.5574,[79]3.5950,
475
+ save_imatrix: stored collected data after 80 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
476
+ [80]3.7128,[81]3.7851,[82]3.7799,[83]3.7675,[84]3.7440,[85]3.8345,[86]3.8787,[87]3.8800,[88]3.8999,[89]3.9499,
477
+ save_imatrix: stored collected data after 90 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
478
+ [90]4.0183,[91]4.0139,[92]4.0121,[93]4.0164,[94]4.0088,[95]3.9814,[96]4.0118,[97]4.0338,[98]4.0638,[99]4.0308,
479
+ save_imatrix: stored collected data after 100 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
480
+ [100]4.0811,[101]4.1422,[102]4.1939,[103]4.2519,[104]4.2992,[105]4.3474,[106]4.3983,[107]4.3891,[108]4.3950,[109]4.4223,
481
+ save_imatrix: stored collected data after 110 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
482
+ [110]4.4616,[111]4.4628,[112]4.5131,[113]4.5552,[114]4.5702,[115]4.5605,[116]4.5330,[117]4.5458,[118]4.5474,[119]4.5088,
483
+ save_imatrix: stored collected data after 120 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
484
+ [120]4.4704,[121]4.4584,[122]4.4520,[123]4.4576,[124]4.4923,[125]4.4902,[126]4.5301,[127]4.5808,[128]4.6256,[129]4.5970,
485
+ save_imatrix: stored collected data after 130 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
486
+ [130]4.5674,[131]4.5513,[132]4.5352,[133]4.5398,[134]4.5298,[135]4.5768,[136]4.6216,[137]4.6465,[138]4.6503,[139]4.6840,
487
+ save_imatrix: stored collected data after 140 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
488
+ [140]4.7355,[141]4.7856,[142]4.8292,[143]4.8613,[144]4.8905,[145]4.9105,[146]4.9111,[147]4.9161,[148]4.9055,[149]4.9285,
489
+ save_imatrix: stored collected data after 150 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
490
+ [150]4.9394,[151]4.9498,[152]4.9741,[153]5.0031,[154]5.0031,[155]5.0045,[156]5.0191,[157]5.0347,[158]5.0440,[159]5.0578,
491
+ save_imatrix: stored collected data after 160 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
492
+ [160]5.0635,[161]5.0710,[162]5.0824,[163]5.0815,[164]5.0738,[165]5.1019,[166]5.1153,[167]5.1173,[168]5.1427,[169]5.1669,
493
+ save_imatrix: stored collected data after 170 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
494
+ [170]5.1473,[171]5.1656,[172]5.1706,[173]5.1913,[174]5.2133,[175]5.2240,[176]5.2127,[177]5.1976,[178]5.1853,[179]5.1704,
495
+ save_imatrix: stored collected data after 180 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
496
+ [180]5.1552,[181]5.1410,[182]5.1326,[183]5.1509,[184]5.1726,[185]5.2238,[186]5.2688,[187]5.3029,[188]5.3528,[189]5.3781,
497
+ save_imatrix: stored collected data after 190 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
498
+ [190]5.4022,[191]5.3836,[192]5.4077,[193]5.3986,[194]5.3637,[195]5.3234,[196]5.3419,[197]5.3768,[198]5.3905,[199]5.4038,
499
+ save_imatrix: stored collected data after 200 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
500
+ [200]5.4335,[201]5.4573,[202]5.4781,[203]5.5035,[204]5.5272,[205]5.5297,[206]5.5102,[207]5.4930,[208]5.4923,[209]5.4749,
501
+ save_imatrix: stored collected data after 210 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
502
+ [210]5.4644,[211]5.4467,[212]5.4250,[213]5.4277,[214]5.4415,[215]5.4205,[216]5.4175,[217]5.4225,[218]5.4355,[219]5.4566,
503
+ save_imatrix: stored collected data after 220 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
504
+ [220]5.4845,[221]5.5102,[222]5.5357,[223]5.5438,[224]5.5872,[225]5.6260,[226]5.6398,[227]5.6421,[228]5.6588,[229]5.6900,
505
+ save_imatrix: stored collected data after 230 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
506
+ [230]5.7133,[231]5.7253,[232]5.7605,[233]5.7680,[234]5.8151,[235]5.8574,[236]5.8745,[237]5.8959,[238]5.9191,[239]5.9352,
507
+ save_imatrix: stored collected data after 240 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
508
+ [240]5.9543,[241]5.9810,[242]6.0051,[243]6.0283,[244]6.0441,[245]6.0680,[246]6.0936,[247]6.1145,[248]6.1212,[249]6.1335,
509
+ save_imatrix: stored collected data after 250 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
510
+ [250]6.1408,[251]6.1578,[252]6.1708,[253]6.1907,[254]6.2048,[255]6.2183,[256]6.2155,[257]6.2234,[258]6.2454,[259]6.2756,
511
+ save_imatrix: stored collected data after 260 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
512
+ [260]6.3059,[261]6.3243,[262]6.3548,[263]6.3571,[264]6.3698,[265]6.3824,[266]6.3960,[267]6.4167,[268]6.4371,[269]6.4558,
513
+ save_imatrix: stored collected data after 270 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
514
+ [270]6.4693,[271]6.4706,[272]6.4964,[273]6.5114,[274]6.5362,[275]6.5569,[276]6.5564,[277]6.5617,[278]6.5767,[279]6.5802,
515
+ save_imatrix: stored collected data after 280 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
516
+ [280]6.5975,[281]6.6092,[282]6.6280,[283]6.6407,[284]6.6626,[285]6.6823,[286]6.7023,[287]6.7206,[288]6.7522,[289]6.7722,
517
+ save_imatrix: stored collected data after 290 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
518
+ [290]6.7935,[291]6.8167,[292]6.8329,[293]6.8436,[294]6.8604,[295]6.8677,[296]6.8786,[297]6.8961,[298]6.9088,[299]6.9215,
519
+ save_imatrix: stored collected data after 300 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
520
+ [300]6.9300,[301]6.9391,[302]6.9487,[303]6.9748,[304]6.9890,[305]6.9988,[306]7.0184,[307]7.0456,[308]7.0756,[309]7.1029,
521
+ save_imatrix: stored collected data after 310 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
522
+ [310]7.0595,[311]7.0437,[312]7.0151,[313]6.9929,[314]7.0202,[315]7.0333,[316]7.0051,[317]7.0187,[318]7.0321,[319]7.0287,
523
+ save_imatrix: stored collected data after 320 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
524
+ [320]7.0292,[321]7.0347,[322]7.0606,[323]7.0596,[324]7.0731,[325]7.0907,[326]7.1053,[327]7.1201,[328]7.0776,[329]7.0961,
525
+ save_imatrix: stored collected data after 330 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
526
+ [330]7.1338,[331]7.1613,[332]7.1877,[333]7.2135,[334]7.2131,[335]7.2100,[336]7.2196,[337]7.2257,[338]7.2430,[339]7.2647,
527
+ save_imatrix: stored collected data after 340 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
528
+ [340]7.2429,[341]7.2588,[342]7.2679,[343]7.2644,[344]7.2647,[345]7.2656,[346]7.2518,[347]7.2605,[348]7.2791,[349]7.2685,
529
+ save_imatrix: stored collected data after 350 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
530
+ [350]7.2599,[351]7.2307,[352]7.1949,[353]7.1699,[354]7.1475,[355]7.1109,[356]7.0880,[357]7.0686,[358]7.0488,[359]7.0303,
531
+ save_imatrix: stored collected data after 360 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
532
+ [360]7.0089,[361]6.9886,[362]6.9868,[363]6.9704,[364]6.9412,[365]6.9239,[366]6.8979,[367]6.8821,[368]6.8681,[369]6.8420,
533
+ save_imatrix: stored collected data after 370 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
534
+ [370]6.8320,[371]6.8266,[372]6.8237,[373]6.8128,[374]6.7941,[375]6.7615,[376]6.7304,[377]6.7149,[378]6.6930,[379]6.6661,
535
+ save_imatrix: stored collected data after 380 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
536
+ [380]6.6338,[381]6.6010,[382]6.5842,[383]6.5753,[384]6.5680,[385]6.5561,[386]6.5725,[387]6.5692,[388]6.5422,[389]6.5191,
537
+ save_imatrix: stored collected data after 390 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
538
+ [390]6.5028,[391]6.4825,[392]6.4643,[393]6.4496,[394]6.4257,[395]6.4064,[396]6.3898,[397]6.3681,[398]6.3439,[399]6.3203,
539
+ save_imatrix: stored collected data after 400 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
540
+ [400]6.3079,[401]6.2973,[402]6.2837,[403]6.2782,[404]6.2749,[405]6.2642,[406]6.2559,[407]6.2414,[408]6.2154,[409]6.1894,
541
+ save_imatrix: stored collected data after 410 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
542
+ [410]6.1638,[411]6.1418,[412]6.1162,[413]6.0923,[414]6.0714,[415]6.0458,[416]6.0250,[417]6.0073,[418]5.9871,[419]5.9676,
543
+ save_imatrix: stored collected data after 420 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
544
+ [420]5.9470,[421]5.9249,[422]5.9074,[423]5.9091,[424]5.9010,[425]5.8958,[426]5.8760,[427]5.8586,[428]5.8375,[429]5.8194,
545
+ save_imatrix: stored collected data after 430 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
546
+ [430]5.8022,[431]5.7885,[432]5.7719,[433]5.7611,[434]5.7449,[435]5.7391,[436]5.7237,[437]5.7058,[438]5.7005,[439]5.6861,
547
+ save_imatrix: stored collected data after 440 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
548
+ [440]5.6712,[441]5.6568,[442]5.6466,[443]5.6298,[444]5.6099,[445]5.5903,[446]5.5717,[447]5.5519,[448]5.5326,[449]5.5163,
549
+ save_imatrix: stored collected data after 450 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
550
+ [450]5.4969,[451]5.4781,[452]5.4617,[453]5.4562,[454]5.4372,[455]5.4286,[456]5.4145,[457]5.4103,[458]5.3989,[459]5.3871,
551
+ save_imatrix: stored collected data after 460 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
552
+ [460]5.3765,[461]5.3653,[462]5.3547,[463]5.3435,[464]5.3327,[465]5.3219,[466]5.3112,[467]5.3002,[468]5.2927,[469]5.2821,
553
+ save_imatrix: stored collected data after 470 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
554
+ [470]5.2688,[471]5.2511,[472]5.2386,[473]5.2345,[474]5.2388,[475]5.2207,[476]5.2076,[477]5.1938,[478]5.1777,[479]5.1635,
555
+ save_imatrix: stored collected data after 480 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
556
+ [480]5.1486,[481]5.1391,[482]5.1274,[483]5.1208,[484]5.1108,[485]5.1004,[486]5.0933,[487]5.0843,[488]5.0738,[489]5.0712,
557
+ save_imatrix: stored collected data after 490 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
558
+ [490]5.0715,[491]5.0750,[492]5.0731,[493]5.0748,[494]5.0754,[495]5.0720,[496]5.0647,[497]5.0758,[498]5.0891,[499]5.1042,
559
+ save_imatrix: stored collected data after 500 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
560
+ [500]5.1145,[501]5.1254,[502]5.1377,[503]5.1500,[504]5.1586,[505]5.1751,[506]5.1879,[507]5.1988,[508]5.2218,[509]5.2465,
561
+ save_imatrix: stored collected data after 510 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
562
+ [510]5.2716,[511]5.2686,[512]5.2800,[513]5.2934,[514]5.3038,[515]5.3089,[516]5.3158,[517]5.3198,[518]5.3211,[519]5.3273,
563
+ save_imatrix: stored collected data after 520 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
564
+ [520]5.3322,[521]5.3420,[522]5.3407,[523]5.3395,[524]5.3485,[525]5.3749,[526]5.3980,[527]5.3991,[528]5.4007,[529]5.4041,
565
+ save_imatrix: stored collected data after 530 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
566
+ [530]5.4079,[531]5.4105,[532]5.4148,[533]5.4181,[534]5.4237,[535]5.4300,[536]5.4400,[537]5.4535,[538]5.4647,[539]5.4754,
567
+ save_imatrix: stored collected data after 540 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
568
+ [540]5.4833,[541]5.4879,[542]5.4937,[543]5.4936,[544]5.4891,[545]5.4841,[546]5.4875,[547]5.4898,[548]5.4953,[549]5.4987,
569
+ save_imatrix: stored collected data after 550 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
570
+ [550]5.5043,[551]5.5035,[552]5.5094,[553]5.5107,[554]5.5154,[555]5.5127,[556]5.5156,[557]5.5126,[558]5.5058,[559]5.5026,
571
+ save_imatrix: stored collected data after 560 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
572
+ [560]5.4951,[561]5.4855,[562]5.4787,[563]5.4741,[564]5.4685,[565]5.4666,[566]5.4657,[567]5.4612,[568]5.4647,[569]5.4620,
573
+ save_imatrix: stored collected data after 570 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
574
+ [570]5.4593,[571]5.4565,[572]5.4557,[573]5.4501,[574]5.4500,[575]5.4437,[576]5.4366,[577]5.4354,[578]5.4349,[579]5.4221,
575
+ save_imatrix: stored collected data after 580 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
576
+ [580]5.4108,[581]5.3990,[582]5.3873,[583]5.3841,[584]5.3858,[585]5.3878,[586]5.3844,[587]5.3816,[588]5.3760,[589]5.3742,
577
+ save_imatrix: stored collected data after 590 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
578
+ [590]5.3760,[591]5.3727,[592]5.3678,[593]5.3653,[594]5.3627,[595]5.3625,[596]5.3602,[597]5.3576,[598]5.3524,[599]5.3527,
579
+ save_imatrix: stored collected data after 600 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
580
+ [600]5.3397,[601]5.3306,[602]5.3214,[603]5.3112,[604]5.3034,[605]5.2952,[606]5.2832,[607]5.2688,[608]5.2544,[609]5.2403,
581
+ save_imatrix: stored collected data after 610 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
582
+ [610]5.2380,[611]5.2296,[612]5.2281,[613]5.2184,[614]5.2166,[615]5.2094,[616]5.2129,[617]5.2115,[618]5.2185,[619]5.2136,
583
+ save_imatrix: stored collected data after 620 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
584
+ [620]5.2109,[621]5.2041,[622]5.2001,[623]5.1963,[624]5.1912,[625]5.1884,[626]5.1852,[627]5.1806,[628]5.1794,[629]5.1857,
585
+ save_imatrix: stored collected data after 630 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
586
+ [630]5.1878,[631]5.1874,[632]5.1867,[633]5.1888,[634]5.1905,[635]5.1918,[636]5.1956,[637]5.1966,[638]5.1964,[639]5.1990,
587
+ save_imatrix: stored collected data after 640 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
588
+ [640]5.1947,[641]5.1989,[642]5.2005,[643]5.2037,[644]5.2043,[645]5.2043,[646]5.2063,[647]5.2032,[648]5.1926,[649]5.1865,
589
+ save_imatrix: stored collected data after 650 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
590
+ [650]5.1788,[651]5.1681,[652]5.1570,[653]5.1523,[654]5.1478,[655]5.1419,[656]5.1315,[657]5.1249,[658]5.1194,[659]5.1122,
591
+ save_imatrix: stored collected data after 660 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
592
+ [660]5.1026,[661]5.0926,[662]5.0856,[663]5.0767,[664]5.0729,[665]5.0666,[666]5.0593,[667]5.0497,[668]5.0473,[669]5.0400,
593
+ save_imatrix: stored collected data after 670 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
594
+ [670]5.0358,[671]5.0312,[672]5.0276,[673]5.0215,[674]5.0113,[675]5.0023,[676]4.9960,[677]4.9874,[678]4.9790,[679]4.9751,
595
+ save_imatrix: stored collected data after 680 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
596
+ [680]4.9707,[681]4.9679,[682]4.9652,[683]4.9612,[684]4.9580,[685]4.9531,[686]4.9502,[687]4.9484,[688]4.9451,[689]4.9410,
597
+ save_imatrix: stored collected data after 690 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
598
+ [690]4.9371,[691]4.9344,[692]4.9317,[693]4.9281,[694]4.9253,[695]4.9205,[696]4.9239,[697]4.9236,[698]4.9250,[699]4.9253,
599
+ save_imatrix: stored collected data after 700 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
600
+ [700]4.9264,[701]4.9266,[702]4.9274,[703]4.9282,[704]4.9293,[705]4.9305,[706]4.9297,[707]4.9306,[708]4.9321,[709]4.9341,
601
+ save_imatrix: stored collected data after 710 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
602
+ [710]4.9370,[711]4.9386,[712]4.9391,[713]4.9395,[714]4.9398,[715]4.9423,[716]4.9440,[717]4.9440,[718]4.9427,[719]4.9408,
603
+ save_imatrix: stored collected data after 720 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
604
+ [720]4.9421,[721]4.9446,[722]4.9459,[723]4.9476,[724]4.9484,[725]4.9493,[726]4.9485,[727]4.9486,[728]4.9493,[729]4.9506,
605
+ save_imatrix: stored collected data after 730 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
606
+ [730]4.9509,[731]4.9526,[732]4.9524,[733]4.9536,[734]4.9544,[735]4.9562,[736]4.9564,[737]4.9573,[738]4.9576,[739]4.9599,
607
+ save_imatrix: stored collected data after 740 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
608
+ [740]4.9624,[741]4.9633,[742]4.9636,[743]4.9641,[744]4.9640,[745]4.9665,[746]4.9665,[747]4.9676,[748]4.9677,[749]4.9695,
609
+ save_imatrix: stored collected data after 750 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
610
+ [750]4.9703,[751]4.9701,[752]4.9708,[753]4.9718,[754]4.9729,[755]4.9751,[756]4.9740,[757]4.9740,[758]4.9761,[759]4.9784,
611
+ save_imatrix: stored collected data after 760 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
612
+ [760]4.9807,[761]4.9807,[762]4.9810,[763]4.9809,[764]4.9803,[765]4.9808,[766]4.9811,[767]4.9815,[768]4.9813,[769]4.9834,
613
+ save_imatrix: stored collected data after 770 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
614
+ [770]4.9832,[771]4.9852,[772]4.9855,[773]4.9849,[774]4.9859,[775]4.9868,[776]4.9901,[777]4.9946,[778]4.9947,[779]4.9945,
615
+ save_imatrix: stored collected data after 780 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
616
+ [780]4.9944,[781]4.9979,[782]4.9989,[783]4.9984,[784]4.9997,[785]5.0025,[786]5.0034,[787]5.0038,[788]5.0037,[789]5.0039,
617
+ save_imatrix: stored collected data after 790 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
618
+ [790]5.0040,[791]5.0038,[792]5.0053,[793]5.0062,[794]5.0060,[795]5.0065,[796]5.0077,
619
+ save_imatrix: stored collected data after 796 chunks in /mnt/data/models/ubergarm/MiniMax-M2.7-GGUF/imatrix-MiniMax-M2.7-BF16.dat
620
+
621
+ Final estimate: PPL = 5.0077 +/- 0.02927
622
+
623
+ ======================== sorted layer importances
624
+ 0: Layer 0, <cos_sim> = 0.314319
625
+ 1: Layer 61, <cos_sim> = 0.747689
626
+ 2: Layer 1, <cos_sim> = 0.756628
627
+ 3: Layer 4, <cos_sim> = 0.83165
628
+ 4: Layer 5, <cos_sim> = 0.85227
629
+ 5: Layer 3, <cos_sim> = 0.879032
630
+ 6: Layer 2, <cos_sim> = 0.913412
631
+ 7: Layer 60, <cos_sim> = 0.921196
632
+ 8: Layer 6, <cos_sim> = 0.929439
633
+ 9: Layer 27, <cos_sim> = 0.929467
634
+ 10: Layer 24, <cos_sim> = 0.936035
635
+ 11: Layer 31, <cos_sim> = 0.936311
636
+ 12: Layer 7, <cos_sim> = 0.936849
637
+ 13: Layer 28, <cos_sim> = 0.937045
638
+ 14: Layer 23, <cos_sim> = 0.939549
639
+ 15: Layer 8, <cos_sim> = 0.94017
640
+ 16: Layer 26, <cos_sim> = 0.940291
641
+ 17: Layer 39, <cos_sim> = 0.942122
642
+ 18: Layer 32, <cos_sim> = 0.942581
643
+ 19: Layer 9, <cos_sim> = 0.94394
644
+ 20: Layer 25, <cos_sim> = 0.944352
645
+ 21: Layer 30, <cos_sim> = 0.945021
646
+ 22: Layer 29, <cos_sim> = 0.945598
647
+ 23: Layer 37, <cos_sim> = 0.946253
648
+ 24: Layer 18, <cos_sim> = 0.947922
649
+ 25: Layer 38, <cos_sim> = 0.949
650
+ 26: Layer 11, <cos_sim> = 0.949036
651
+ 27: Layer 34, <cos_sim> = 0.949104
652
+ 28: Layer 41, <cos_sim> = 0.949253
653
+ 29: Layer 17, <cos_sim> = 0.949477
654
+ 30: Layer 22, <cos_sim> = 0.950325
655
+ 31: Layer 10, <cos_sim> = 0.950418
656
+ 32: Layer 35, <cos_sim> = 0.950613
657
+ 33: Layer 58, <cos_sim> = 0.950771
658
+ 34: Layer 15, <cos_sim> = 0.951046
659
+ 35: Layer 59, <cos_sim> = 0.95122
660
+ 36: Layer 49, <cos_sim> = 0.951834
661
+ 37: Layer 43, <cos_sim> = 0.952029
662
+ 38: Layer 16, <cos_sim> = 0.952079
663
+ 39: Layer 36, <cos_sim> = 0.95235
664
+ 40: Layer 40, <cos_sim> = 0.952539
665
+ 41: Layer 21, <cos_sim> = 0.952563
666
+ 42: Layer 12, <cos_sim> = 0.953459
667
+ 43: Layer 46, <cos_sim> = 0.954189
668
+ 44: Layer 33, <cos_sim> = 0.955066
669
+ 45: Layer 57, <cos_sim> = 0.955185
670
+ 46: Layer 13, <cos_sim> = 0.955958
671
+ 47: Layer 19, <cos_sim> = 0.958471
672
+ 48: Layer 48, <cos_sim> = 0.959151
673
+ 49: Layer 54, <cos_sim> = 0.960254
674
+ 50: Layer 51, <cos_sim> = 0.960338
675
+ 51: Layer 44, <cos_sim> = 0.960788
676
+ 52: Layer 42, <cos_sim> = 0.960864
677
+ 53: Layer 45, <cos_sim> = 0.961023
678
+ 54: Layer 50, <cos_sim> = 0.961289
679
+ 55: Layer 56, <cos_sim> = 0.962136
680
+ 56: Layer 20, <cos_sim> = 0.962164
681
+ 57: Layer 52, <cos_sim> = 0.962365
682
+ 58: Layer 47, <cos_sim> = 0.963786
683
+ 59: Layer 53, <cos_sim> = 0.96446
684
+ 60: Layer 55, <cos_sim> = 0.964589
685
+ 61: Layer 14, <cos_sim> = 0.964797
686
+
687
+ ======================== sorted attention importances
688
+ 0: Layer 0, <cos_sim> = 0.070671
689
+ 1: Layer 5, <cos_sim> = 0.325468
690
+ 2: Layer 3, <cos_sim> = 0.369615
691
+ 3: Layer 2, <cos_sim> = 0.412956
692
+ 4: Layer 4, <cos_sim> = 0.573739
693
+ 5: Layer 7, <cos_sim> = 0.585959
694
+ 6: Layer 11, <cos_sim> = 0.586286
695
+ 7: Layer 6, <cos_sim> = 0.595757
696
+ 8: Layer 9, <cos_sim> = 0.607441
697
+ 9: Layer 12, <cos_sim> = 0.655411
698
+ 10: Layer 8, <cos_sim> = 0.666884
699
+ 11: Layer 15, <cos_sim> = 0.668835
700
+ 12: Layer 10, <cos_sim> = 0.66924
701
+ 13: Layer 17, <cos_sim> = 0.685496
702
+ 14: Layer 1, <cos_sim> = 0.695764
703
+ 15: Layer 14, <cos_sim> = 0.696466
704
+ 16: Layer 22, <cos_sim> = 0.708887
705
+ 17: Layer 20, <cos_sim> = 0.73429
706
+ 18: Layer 23, <cos_sim> = 0.734495
707
+ 19: Layer 16, <cos_sim> = 0.738226
708
+ 20: Layer 21, <cos_sim> = 0.740639
709
+ 21: Layer 18, <cos_sim> = 0.742008
710
+ 22: Layer 19, <cos_sim> = 0.761501
711
+ 23: Layer 13, <cos_sim> = 0.762193
712
+ 24: Layer 26, <cos_sim> = 0.792264
713
+ 25: Layer 24, <cos_sim> = 0.801974
714
+ 26: Layer 27, <cos_sim> = 0.811254
715
+ 27: Layer 28, <cos_sim> = 0.822608
716
+ 28: Layer 25, <cos_sim> = 0.823736
717
+ 29: Layer 29, <cos_sim> = 0.841409
718
+ 30: Layer 30, <cos_sim> = 0.848063
719
+ 31: Layer 31, <cos_sim> = 0.862063
720
+ 32: Layer 32, <cos_sim> = 0.86227
721
+ 33: Layer 33, <cos_sim> = 0.869472
722
+ 34: Layer 34, <cos_sim> = 0.881373
723
+ 35: Layer 36, <cos_sim> = 0.884664
724
+ 36: Layer 35, <cos_sim> = 0.893423
725
+ 37: Layer 37, <cos_sim> = 0.894579
726
+ 38: Layer 39, <cos_sim> = 0.89667
727
+ 39: Layer 38, <cos_sim> = 0.896756
728
+ 40: Layer 61, <cos_sim> = 0.899749
729
+ 41: Layer 40, <cos_sim> = 0.904017
730
+ 42: Layer 43, <cos_sim> = 0.922443
731
+ 43: Layer 41, <cos_sim> = 0.927151
732
+ 44: Layer 42, <cos_sim> = 0.93657
733
+ 45:
734
+ llama_print_timings: load time = 95309.51 ms
735
+ llama_print_timings: sample time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second)
736
+ llama_print_timings: prompt eval time = 3029558.42 ms / 407552 tokens ( 7.43 ms per token, 134.53 tokens per second)
737
+ llama_print_timings: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second)
738
+ llama_print_timings: total time = 3135252.92 ms / 407553 tokens
739
+ Layer 49, <cos_sim> = 0.941207
740
+ 46: Layer 44, <cos_sim> = 0.94793
741
+ 47: Layer 45, <cos_sim> = 0.948874
742
+ 48: Layer 46, <cos_sim> = 0.951086
743
+ 49: Layer 47, <cos_sim> = 0.954018
744
+ 50: Layer 52, <cos_sim> = 0.955026
745
+ 51: Layer 48, <cos_sim> = 0.960921
746
+ 52: Layer 50, <cos_sim> = 0.96179
747
+ 53: Layer 60, <cos_sim> = 0.962736
748
+ 54: Layer 58, <cos_sim> = 0.965627
749
+ 55: Layer 54, <cos_sim> = 0.965685
750
+ 56: Layer 51, <cos_sim> = 0.966247
751
+ 57: Layer 59, <cos_sim> = 0.969823
752
+ 58: Layer 53, <cos_sim> = 0.970095
753
+ 59: Layer 57, <cos_sim> = 0.970563
754
+ 60: Layer 56, <cos_sim> = 0.972371
755
+ 61: Layer 55, <cos_sim> = 0.974544
756
+
757
+ ======================== sorted ffn importances
758
+ 0: Layer 4, <cos_sim> = 0.317384
759
+ 1: Layer 2, <cos_sim> = 0.460713
760
+ 2: Layer 3, <cos_sim> = 0.566183
761
+ 3: Layer 6, <cos_sim> = 0.57249
762
+ 4: Layer 5, <cos_sim> = 0.579167
763
+ 5: Layer 10, <cos_sim> = 0.601761
764
+ 6: Layer 8, <cos_sim> = 0.601878
765
+ 7: Layer 0, <cos_sim> = 0.622948
766
+ 8: Layer 11, <cos_sim> = 0.638515
767
+ 9: Layer 9, <cos_sim> = 0.6441
768
+ 10: Layer 14, <cos_sim> = 0.649876
769
+ 11: Layer 7, <cos_sim> = 0.670979
770
+ 12: Layer 16, <cos_sim> = 0.679592
771
+ 13: Layer 13, <cos_sim> = 0.684167
772
+ 14: Layer 21, <cos_sim> = 0.689835
773
+ 15: Layer 1, <cos_sim> = 0.709895
774
+ 16: Layer 19, <cos_sim> = 0.710139
775
+ 17: Layer 20, <cos_sim> = 0.721436
776
+ 18: Layer 22, <cos_sim> = 0.7316
777
+ 19: Layer 15, <cos_sim> = 0.732324
778
+ 20: Layer 17, <cos_sim> = 0.742242
779
+ 21: Layer 12, <cos_sim> = 0.742858
780
+ 22: Layer 18, <cos_sim> = 0.746992
781
+ 23: Layer 25, <cos_sim> = 0.779369
782
+ 24: Layer 23, <cos_sim> = 0.781141
783
+ 25: Layer 26, <cos_sim> = 0.796414
784
+ 26: Layer 24, <cos_sim> = 0.809111
785
+ 27: Layer 27, <cos_sim> = 0.810032
786
+ 28: Layer 29, <cos_sim> = 0.821859
787
+ 29: Layer 28, <cos_sim> = 0.824159
788
+ 30: Layer 30, <cos_sim> = 0.843586
789
+ 31: Layer 31, <cos_sim> = 0.848297
790
+ 32: Layer 32, <cos_sim> = 0.855983
791
+ 33: Layer 33, <cos_sim> = 0.86832
792
+ 34: Layer 35, <cos_sim> = 0.871202
793
+ 35: Layer 61, <cos_sim> = 0.872646
794
+ 36: Layer 36, <cos_sim> = 0.883183
795
+ 37: Layer 34, <cos_sim> = 0.88452
796
+ 38: Layer 39, <cos_sim> = 0.888821
797
+ 39: Layer 37, <cos_sim> = 0.892246
798
+ 40: Layer 38, <cos_sim> = 0.895153
799
+ 41: Layer 42, <cos_sim> = 0.907247
800
+ 42: Layer 40, <cos_sim> = 0.908511
801
+ 43: Layer 41, <cos_sim> = 0.916338
802
+ 44: Layer 48, <cos_sim> = 0.926847
803
+ 45: Layer 44, <cos_sim> = 0.927916
804
+ 46: Layer 43, <cos_sim> = 0.931278
805
+ 47: Layer 46, <cos_sim> = 0.933581
806
+ 48: Layer 51, <cos_sim> = 0.937163
807
+ 49: Layer 45, <cos_sim> = 0.937349
808
+ 50: Layer 49, <cos_sim> = 0.939469
809
+ 51: Layer 47, <cos_sim> = 0.943874
810
+ 52: Layer 53, <cos_sim> = 0.947779
811
+ 53: Layer 50, <cos_sim> = 0.947821
812
+ 54: Layer 54, <cos_sim> = 0.951103
813
+ 55: Layer 52, <cos_sim> = 0.951138
814
+ 56: Layer 56, <cos_sim> = 0.955928
815
+ 57: Layer 55, <cos_sim> = 0.956456
816
+ 58: Layer 60, <cos_sim> = 0.957153
817
+ 59: Layer 57, <cos_sim> = 0.962098
818
+ 60: Layer 58, <cos_sim> = 0.970282
819
+ 61: Layer 59, <cos_sim> = 0.971508