Hugging Face's logo

praison
/
axolotl-openllama-1k-qloras

PEFT
TensorBoard
Safetensors
llama
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files
xet
Metrics Community
praison commited on
Commit
eb1c88e
·
verified ·
1 Parent(s): d9db993

Upload folder using huggingface_hub

Browse files
Files changed (17) hide show
  1. README.md +135 -0
  2. adapter_config.json +33 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-431/README.md +202 -0
  5. checkpoint-431/adapter_config.json +33 -0
  6. checkpoint-431/adapter_model.safetensors +3 -0
  7. checkpoint-431/optimizer.pt +3 -0
  8. checkpoint-431/rng_state.pth +3 -0
  9. checkpoint-431/scheduler.pt +3 -0
  10. checkpoint-431/trainer_state.json +3070 -0
  11. checkpoint-431/training_args.bin +3 -0
  12. config.json +43 -0
  13. runs/Mar08_18-22-13_80d2e4e4a3f6/events.out.tfevents.1709922133.80d2e4e4a3f6.198.0 +3 -0
  14. runs/Mar08_18-28-26_80d2e4e4a3f6/events.out.tfevents.1709922507.80d2e4e4a3f6.226.0 +3 -0
  15. special_tokens_map.json +24 -0
  16. tokenizer.model +3 -0
  17. tokenizer_config.json +43 -0
README.md ADDED
@@ -0,0 +1,135 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: openlm-research/open_llama_3b_v2
7
+ model-index:
8
+ - name: qlora-out
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.4.0`
19
+ ```yaml
20
+ base_model: openlm-research/open_llama_3b_v2
21
+ model_type: LlamaForCausalLM
22
+ tokenizer_type: LlamaTokenizer
23
+ load_in_8bit: false
24
+ load_in_4bit: true
25
+ strict: false
26
+ push_dataset_to_hub:
27
+ datasets:
28
+ - path: mhenrichsen/alpaca_2k_test
29
+ type: alpaca
30
+ dataset_prepared_path:
31
+ val_set_size: 0.05
32
+ adapter: qlora
33
+ lora_model_dir:
34
+ sequence_len: 1024
35
+ sample_packing: true
36
+ lora_r: 32
37
+ lora_alpha: 32
38
+ lora_dropout: 0.05
39
+ lora_target_modules:
40
+ lora_target_linear: true
41
+ lora_fan_in_fan_out:
42
+ wandb_project:
43
+ wandb_entity:
44
+ wandb_watch:
45
+ wandb_name:
46
+ wandb_log_model:
47
+ output_dir: ./qlora-out
48
+ gradient_accumulation_steps: 1
49
+ micro_batch_size: 1
50
+ num_epochs: 1
51
+ optimizer: paged_adamw_32bit
52
+ torchdistx_path:
53
+ lr_scheduler: cosine
54
+ learning_rate: 0.0002
55
+ train_on_inputs: false
56
+ group_by_length: false
57
+ bf16: false
58
+ fp16: true
59
+ tf32: false
60
+ gradient_checkpointing: true
61
+ early_stopping_patience:
62
+ resume_from_checkpoint:
63
+ local_rank:
64
+ logging_steps: 1
65
+ xformers_attention:
66
+ flash_attention: true
67
+ gptq_groupsize:
68
+ gptq_model_v1:
69
+ warmup_steps: 20
70
+ evals_per_epoch: 4
71
+ saves_per_epoch: 1
72
+ debug:
73
+ deepspeed:
74
+ weight_decay: 0.1
75
+ fsdp:
76
+ fsdp_config:
77
+ special_tokens:
78
+ bos_token: "<s>"
79
+ eos_token: "</s>"
80
+ unk_token: "<unk>"
81
+
82
+ ```
83
+
84
+ </details><br>
85
+
86
+ # qlora-out
87
+
88
+ This model is a fine-tuned version of [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2) on the None dataset.
89
+ It achieves the following results on the evaluation set:
90
+ - Loss: 1.1097
91
+
92
+ ## Model description
93
+
94
+ More information needed
95
+
96
+ ## Intended uses & limitations
97
+
98
+ More information needed
99
+
100
+ ## Training and evaluation data
101
+
102
+ More information needed
103
+
104
+ ## Training procedure
105
+
106
+ ### Training hyperparameters
107
+
108
+ The following hyperparameters were used during training:
109
+ - learning_rate: 0.0002
110
+ - train_batch_size: 1
111
+ - eval_batch_size: 1
112
+ - seed: 42
113
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
114
+ - lr_scheduler_type: cosine
115
+ - lr_scheduler_warmup_steps: 20
116
+ - num_epochs: 1
117
+ - mixed_precision_training: Native AMP
118
+
119
+ ### Training results
120
+
121
+ | Training Loss | Epoch | Step | Validation Loss |
122
+ |:-------------:|:-----:|:----:|:---------------:|
123
+ | 1.2567 | 0.0 | 1 | 1.3470 |
124
+ | 1.1738 | 0.25 | 108 | 1.1372 |
125
+ | 1.1175 | 0.5 | 216 | 1.1233 |
126
+ | 1.4102 | 0.75 | 324 | 1.1097 |
127
+
128
+
129
+ ### Framework versions
130
+
131
+ - PEFT 0.9.0
132
+ - Transformers 4.38.2
133
+ - Pytorch 2.1.2+cu118
134
+ - Datasets 2.18.0
135
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "k_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75e500eda8de3f64d573a45d6a6ce4404f136d57981a68eb1e72ea1ea5545fe7
3
+ size 203538938
checkpoint-431/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-431/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "k_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-431/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1979410dc8e9e2db594d1c68b37aca16315a2a6215b17fbf71970b8831fdb870
3
+ size 203456160
checkpoint-431/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b162b0daaa3f5e79cf72346b15884be1eab4c8796f85a4d5875976b7a8dfb478
3
+ size 407040314
checkpoint-431/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e766413e4f3b39c1d0ac620807b3bd3fd4dac79e0a0eed6a4a60c5746642e0a6
3
+ size 14244
checkpoint-431/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23f73a16ac262980457d80b6b9e4834ebbf9e3ee06ac2280222318fcdf9e15a8
3
+ size 1064
checkpoint-431/trainer_state.json ADDED
@@ -0,0 +1,3070 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 108,
6
+ "global_step": 431,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "grad_norm": 0.44969096779823303,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.2567,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "eval_loss": 1.3469510078430176,
21
+ "eval_runtime": 4.7982,
22
+ "eval_samples_per_second": 20.841,
23
+ "eval_steps_per_second": 20.841,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.0,
28
+ "grad_norm": 0.4098110496997833,
29
+ "learning_rate": 2e-05,
30
+ "loss": 1.3328,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 0.01,
35
+ "grad_norm": 0.5452544093132019,
36
+ "learning_rate": 3e-05,
37
+ "loss": 1.6567,
38
+ "step": 3
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "grad_norm": 0.7305347323417664,
43
+ "learning_rate": 4e-05,
44
+ "loss": 1.5499,
45
+ "step": 4
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "grad_norm": 0.34722596406936646,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.4343,
52
+ "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.01,
56
+ "grad_norm": 0.5768171548843384,
57
+ "learning_rate": 6e-05,
58
+ "loss": 1.2677,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.02,
63
+ "grad_norm": 0.49281951785087585,
64
+ "learning_rate": 7e-05,
65
+ "loss": 1.473,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.02,
70
+ "grad_norm": 0.42547014355659485,
71
+ "learning_rate": 8e-05,
72
+ "loss": 1.406,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.02,
77
+ "grad_norm": 0.40852880477905273,
78
+ "learning_rate": 9e-05,
79
+ "loss": 1.2842,
80
+ "step": 9
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "grad_norm": 0.36900222301483154,
85
+ "learning_rate": 0.0001,
86
+ "loss": 1.1442,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.03,
91
+ "grad_norm": 0.40908315777778625,
92
+ "learning_rate": 0.00011000000000000002,
93
+ "loss": 1.4667,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.03,
98
+ "grad_norm": 0.4117198884487152,
99
+ "learning_rate": 0.00012,
100
+ "loss": 0.9759,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.03,
105
+ "grad_norm": 0.6714757084846497,
106
+ "learning_rate": 0.00013000000000000002,
107
+ "loss": 1.5162,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.03,
112
+ "grad_norm": 0.5178409218788147,
113
+ "learning_rate": 0.00014,
114
+ "loss": 1.1519,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.03,
119
+ "grad_norm": 1.1334081888198853,
120
+ "learning_rate": 0.00015000000000000001,
121
+ "loss": 1.2767,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.04,
126
+ "grad_norm": 0.6541109681129456,
127
+ "learning_rate": 0.00016,
128
+ "loss": 1.3227,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.04,
133
+ "grad_norm": 0.5143316984176636,
134
+ "learning_rate": 0.00017,
135
+ "loss": 1.2991,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.04,
140
+ "grad_norm": 0.5004872679710388,
141
+ "learning_rate": 0.00018,
142
+ "loss": 1.3201,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.04,
147
+ "grad_norm": 0.4225166141986847,
148
+ "learning_rate": 0.00019,
149
+ "loss": 1.2054,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.05,
154
+ "grad_norm": 0.8387117385864258,
155
+ "learning_rate": 0.0002,
156
+ "loss": 1.2026,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.05,
161
+ "grad_norm": 0.6876248717308044,
162
+ "learning_rate": 0.00019999707864731247,
163
+ "loss": 1.2311,
164
+ "step": 21
165
+ },
166
+ {
167
+ "epoch": 0.05,
168
+ "grad_norm": 0.6449929475784302,
169
+ "learning_rate": 0.00019998831475993593,
170
+ "loss": 1.3162,
171
+ "step": 22
172
+ },
173
+ {
174
+ "epoch": 0.05,
175
+ "grad_norm": 0.6817747950553894,
176
+ "learning_rate": 0.00019997370884991842,
177
+ "loss": 1.5836,
178
+ "step": 23
179
+ },
180
+ {
181
+ "epoch": 0.06,
182
+ "grad_norm": 0.5388895273208618,
183
+ "learning_rate": 0.0001999532617706403,
184
+ "loss": 1.3085,
185
+ "step": 24
186
+ },
187
+ {
188
+ "epoch": 0.06,
189
+ "grad_norm": 0.48557716608047485,
190
+ "learning_rate": 0.00019992697471676413,
191
+ "loss": 1.3159,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 0.06,
196
+ "grad_norm": 0.531157910823822,
197
+ "learning_rate": 0.00019989484922416502,
198
+ "loss": 1.2498,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.06,
203
+ "grad_norm": 0.47394225001335144,
204
+ "learning_rate": 0.0001998568871698409,
205
+ "loss": 1.1981,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.06,
210
+ "grad_norm": 0.48895126581192017,
211
+ "learning_rate": 0.00019981309077180272,
212
+ "loss": 1.0461,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.07,
217
+ "grad_norm": 0.4990120828151703,
218
+ "learning_rate": 0.00019976346258894503,
219
+ "loss": 1.3558,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.07,
224
+ "grad_norm": 0.6515373587608337,
225
+ "learning_rate": 0.00019970800552089623,
226
+ "loss": 1.2566,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.07,
231
+ "grad_norm": 0.4564642012119293,
232
+ "learning_rate": 0.00019964672280784954,
233
+ "loss": 1.1579,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.07,
238
+ "grad_norm": 0.6385357975959778,
239
+ "learning_rate": 0.00019957961803037326,
240
+ "loss": 1.1824,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.08,
245
+ "grad_norm": 0.5059694051742554,
246
+ "learning_rate": 0.00019950669510920184,
247
+ "loss": 1.0092,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.08,
252
+ "grad_norm": 0.49390971660614014,
253
+ "learning_rate": 0.0001994279583050067,
254
+ "loss": 1.1982,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.08,
259
+ "grad_norm": 0.6610875129699707,
260
+ "learning_rate": 0.00019934341221814739,
261
+ "loss": 1.5833,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.08,
266
+ "grad_norm": 0.5280133485794067,
267
+ "learning_rate": 0.0001992530617884026,
268
+ "loss": 0.9696,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.09,
273
+ "grad_norm": 0.49730122089385986,
274
+ "learning_rate": 0.00019915691229468178,
275
+ "loss": 1.1866,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.09,
280
+ "grad_norm": 0.8253684043884277,
281
+ "learning_rate": 0.00019905496935471658,
282
+ "loss": 1.7239,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.09,
287
+ "grad_norm": 0.4826942980289459,
288
+ "learning_rate": 0.0001989472389247326,
289
+ "loss": 1.5085,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.09,
294
+ "grad_norm": 0.6226486563682556,
295
+ "learning_rate": 0.00019883372729910152,
296
+ "loss": 1.1311,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.1,
301
+ "grad_norm": 0.4536150097846985,
302
+ "learning_rate": 0.0001987144411099731,
303
+ "loss": 1.243,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.1,
308
+ "grad_norm": 0.48488786816596985,
309
+ "learning_rate": 0.000198589387326888,
310
+ "loss": 1.2221,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.1,
315
+ "grad_norm": 0.4422035813331604,
316
+ "learning_rate": 0.00019845857325637031,
317
+ "loss": 1.5296,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.1,
322
+ "grad_norm": 0.47116392850875854,
323
+ "learning_rate": 0.00019832200654150076,
324
+ "loss": 1.1176,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.1,
329
+ "grad_norm": 0.4207026958465576,
330
+ "learning_rate": 0.0001981796951614701,
331
+ "loss": 1.0939,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.11,
336
+ "grad_norm": 0.5403062105178833,
337
+ "learning_rate": 0.00019803164743111302,
338
+ "loss": 1.2371,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.11,
343
+ "grad_norm": 0.6009058952331543,
344
+ "learning_rate": 0.00019787787200042223,
345
+ "loss": 1.3423,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.11,
350
+ "grad_norm": 0.421722412109375,
351
+ "learning_rate": 0.00019771837785404305,
352
+ "loss": 0.9633,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.11,
357
+ "grad_norm": 0.5040818452835083,
358
+ "learning_rate": 0.00019755317431074859,
359
+ "loss": 1.3888,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.12,
364
+ "grad_norm": 0.5417558550834656,
365
+ "learning_rate": 0.0001973822710228951,
366
+ "loss": 1.0181,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.12,
371
+ "grad_norm": 0.5744611024856567,
372
+ "learning_rate": 0.00019720567797585817,
373
+ "loss": 1.3067,
374
+ "step": 51
375
+ },
376
+ {
377
+ "epoch": 0.12,
378
+ "grad_norm": 0.4889126121997833,
379
+ "learning_rate": 0.0001970234054874493,
380
+ "loss": 1.1691,
381
+ "step": 52
382
+ },
383
+ {
384
+ "epoch": 0.12,
385
+ "grad_norm": 0.6123921275138855,
386
+ "learning_rate": 0.0001968354642073129,
387
+ "loss": 1.2479,
388
+ "step": 53
389
+ },
390
+ {
391
+ "epoch": 0.13,
392
+ "grad_norm": 0.5839653611183167,
393
+ "learning_rate": 0.00019664186511630433,
394
+ "loss": 0.8839,
395
+ "step": 54
396
+ },
397
+ {
398
+ "epoch": 0.13,
399
+ "grad_norm": 0.40772297978401184,
400
+ "learning_rate": 0.000196442619525848,
401
+ "loss": 1.2291,
402
+ "step": 55
403
+ },
404
+ {
405
+ "epoch": 0.13,
406
+ "grad_norm": 0.5853259563446045,
407
+ "learning_rate": 0.00019623773907727682,
408
+ "loss": 1.4134,
409
+ "step": 56
410
+ },
411
+ {
412
+ "epoch": 0.13,
413
+ "grad_norm": 0.6176974177360535,
414
+ "learning_rate": 0.0001960272357411517,
415
+ "loss": 1.3261,
416
+ "step": 57
417
+ },
418
+ {
419
+ "epoch": 0.13,
420
+ "grad_norm": 0.4556988477706909,
421
+ "learning_rate": 0.0001958111218165624,
422
+ "loss": 1.1442,
423
+ "step": 58
424
+ },
425
+ {
426
+ "epoch": 0.14,
427
+ "grad_norm": 0.46846720576286316,
428
+ "learning_rate": 0.00019558940993040885,
429
+ "loss": 1.0652,
430
+ "step": 59
431
+ },
432
+ {
433
+ "epoch": 0.14,
434
+ "grad_norm": 0.4391572177410126,
435
+ "learning_rate": 0.00019536211303666323,
436
+ "loss": 1.3619,
437
+ "step": 60
438
+ },
439
+ {
440
+ "epoch": 0.14,
441
+ "grad_norm": 0.4686979651451111,
442
+ "learning_rate": 0.00019512924441561348,
443
+ "loss": 0.5938,
444
+ "step": 61
445
+ },
446
+ {
447
+ "epoch": 0.14,
448
+ "grad_norm": 0.430164635181427,
449
+ "learning_rate": 0.00019489081767308698,
450
+ "loss": 1.2571,
451
+ "step": 62
452
+ },
453
+ {
454
+ "epoch": 0.15,
455
+ "grad_norm": 0.6731412410736084,
456
+ "learning_rate": 0.00019464684673965583,
457
+ "loss": 1.3784,
458
+ "step": 63
459
+ },
460
+ {
461
+ "epoch": 0.15,
462
+ "grad_norm": 0.39849698543548584,
463
+ "learning_rate": 0.0001943973458698229,
464
+ "loss": 0.9475,
465
+ "step": 64
466
+ },
467
+ {
468
+ "epoch": 0.15,
469
+ "grad_norm": 0.5062661170959473,
470
+ "learning_rate": 0.00019414232964118892,
471
+ "loss": 1.1145,
472
+ "step": 65
473
+ },
474
+ {
475
+ "epoch": 0.15,
476
+ "grad_norm": 0.5099257230758667,
477
+ "learning_rate": 0.00019388181295360078,
478
+ "loss": 1.0426,
479
+ "step": 66
480
+ },
481
+ {
482
+ "epoch": 0.16,
483
+ "grad_norm": 0.5760392546653748,
484
+ "learning_rate": 0.00019361581102828095,
485
+ "loss": 1.057,
486
+ "step": 67
487
+ },
488
+ {
489
+ "epoch": 0.16,
490
+ "grad_norm": 0.41587546467781067,
491
+ "learning_rate": 0.0001933443394069383,
492
+ "loss": 1.181,
493
+ "step": 68
494
+ },
495
+ {
496
+ "epoch": 0.16,
497
+ "grad_norm": 0.47386714816093445,
498
+ "learning_rate": 0.00019306741395085976,
499
+ "loss": 1.1613,
500
+ "step": 69
501
+ },
502
+ {
503
+ "epoch": 0.16,
504
+ "grad_norm": 0.39666488766670227,
505
+ "learning_rate": 0.0001927850508399839,
506
+ "loss": 1.1075,
507
+ "step": 70
508
+ },
509
+ {
510
+ "epoch": 0.16,
511
+ "grad_norm": 0.4820801019668579,
512
+ "learning_rate": 0.00019249726657195532,
513
+ "loss": 1.3065,
514
+ "step": 71
515
+ },
516
+ {
517
+ "epoch": 0.17,
518
+ "grad_norm": 0.4606281518936157,
519
+ "learning_rate": 0.00019220407796116098,
520
+ "loss": 1.3073,
521
+ "step": 72
522
+ },
523
+ {
524
+ "epoch": 0.17,
525
+ "grad_norm": 0.7897779941558838,
526
+ "learning_rate": 0.00019190550213774756,
527
+ "loss": 1.162,
528
+ "step": 73
529
+ },
530
+ {
531
+ "epoch": 0.17,
532
+ "grad_norm": 0.42179742455482483,
533
+ "learning_rate": 0.00019160155654662076,
534
+ "loss": 0.7054,
535
+ "step": 74
536
+ },
537
+ {
538
+ "epoch": 0.17,
539
+ "grad_norm": 0.5278844237327576,
540
+ "learning_rate": 0.00019129225894642593,
541
+ "loss": 1.0787,
542
+ "step": 75
543
+ },
544
+ {
545
+ "epoch": 0.18,
546
+ "grad_norm": 0.42007845640182495,
547
+ "learning_rate": 0.00019097762740851061,
548
+ "loss": 1.0423,
549
+ "step": 76
550
+ },
551
+ {
552
+ "epoch": 0.18,
553
+ "grad_norm": 0.5332473516464233,
554
+ "learning_rate": 0.0001906576803158686,
555
+ "loss": 1.1466,
556
+ "step": 77
557
+ },
558
+ {
559
+ "epoch": 0.18,
560
+ "grad_norm": 0.5041698217391968,
561
+ "learning_rate": 0.0001903324363620659,
562
+ "loss": 1.0691,
563
+ "step": 78
564
+ },
565
+ {
566
+ "epoch": 0.18,
567
+ "grad_norm": 0.2976062595844269,
568
+ "learning_rate": 0.0001900019145501484,
569
+ "loss": 0.6296,
570
+ "step": 79
571
+ },
572
+ {
573
+ "epoch": 0.19,
574
+ "grad_norm": 0.4532124400138855,
575
+ "learning_rate": 0.0001896661341915318,
576
+ "loss": 1.4112,
577
+ "step": 80
578
+ },
579
+ {
580
+ "epoch": 0.19,
581
+ "grad_norm": 0.4600175619125366,
582
+ "learning_rate": 0.0001893251149048732,
583
+ "loss": 1.0353,
584
+ "step": 81
585
+ },
586
+ {
587
+ "epoch": 0.19,
588
+ "grad_norm": 0.5047218203544617,
589
+ "learning_rate": 0.00018897887661492474,
590
+ "loss": 1.1204,
591
+ "step": 82
592
+ },
593
+ {
594
+ "epoch": 0.19,
595
+ "grad_norm": 0.5466210246086121,
596
+ "learning_rate": 0.00018862743955136966,
597
+ "loss": 1.4353,
598
+ "step": 83
599
+ },
600
+ {
601
+ "epoch": 0.19,
602
+ "grad_norm": 0.3947051465511322,
603
+ "learning_rate": 0.0001882708242476401,
604
+ "loss": 1.0523,
605
+ "step": 84
606
+ },
607
+ {
608
+ "epoch": 0.2,
609
+ "grad_norm": 0.4432850778102875,
610
+ "learning_rate": 0.00018790905153971758,
611
+ "loss": 0.8657,
612
+ "step": 85
613
+ },
614
+ {
615
+ "epoch": 0.2,
616
+ "grad_norm": 0.4575612246990204,
617
+ "learning_rate": 0.00018754214256491562,
618
+ "loss": 1.0524,
619
+ "step": 86
620
+ },
621
+ {
622
+ "epoch": 0.2,
623
+ "grad_norm": 0.3713025748729706,
624
+ "learning_rate": 0.00018717011876064453,
625
+ "loss": 1.0439,
626
+ "step": 87
627
+ },
628
+ {
629
+ "epoch": 0.2,
630
+ "grad_norm": 0.4547780156135559,
631
+ "learning_rate": 0.0001867930018631592,
632
+ "loss": 1.01,
633
+ "step": 88
634
+ },
635
+ {
636
+ "epoch": 0.21,
637
+ "grad_norm": 0.421790212392807,
638
+ "learning_rate": 0.00018641081390628877,
639
+ "loss": 1.1949,
640
+ "step": 89
641
+ },
642
+ {
643
+ "epoch": 0.21,
644
+ "grad_norm": 0.4341322183609009,
645
+ "learning_rate": 0.00018602357722014964,
646
+ "loss": 1.2932,
647
+ "step": 90
648
+ },
649
+ {
650
+ "epoch": 0.21,
651
+ "grad_norm": 0.3753775358200073,
652
+ "learning_rate": 0.00018563131442984044,
653
+ "loss": 1.0415,
654
+ "step": 91
655
+ },
656
+ {
657
+ "epoch": 0.21,
658
+ "grad_norm": 0.5391212105751038,
659
+ "learning_rate": 0.00018523404845412027,
660
+ "loss": 1.1474,
661
+ "step": 92
662
+ },
663
+ {
664
+ "epoch": 0.22,
665
+ "grad_norm": 0.7697898149490356,
666
+ "learning_rate": 0.0001848318025040697,
667
+ "loss": 1.3765,
668
+ "step": 93
669
+ },
670
+ {
671
+ "epoch": 0.22,
672
+ "grad_norm": 0.39486926794052124,
673
+ "learning_rate": 0.00018442460008173445,
674
+ "loss": 1.0153,
675
+ "step": 94
676
+ },
677
+ {
678
+ "epoch": 0.22,
679
+ "grad_norm": 0.4511030912399292,
680
+ "learning_rate": 0.0001840124649787524,
681
+ "loss": 1.0658,
682
+ "step": 95
683
+ },
684
+ {
685
+ "epoch": 0.22,
686
+ "grad_norm": 0.5483732223510742,
687
+ "learning_rate": 0.0001835954212749632,
688
+ "loss": 1.1011,
689
+ "step": 96
690
+ },
691
+ {
692
+ "epoch": 0.23,
693
+ "grad_norm": 0.3995431959629059,
694
+ "learning_rate": 0.0001831734933370019,
695
+ "loss": 0.8759,
696
+ "step": 97
697
+ },
698
+ {
699
+ "epoch": 0.23,
700
+ "grad_norm": 0.4208924472332001,
701
+ "learning_rate": 0.0001827467058168748,
702
+ "loss": 0.8876,
703
+ "step": 98
704
+ },
705
+ {
706
+ "epoch": 0.23,
707
+ "grad_norm": 0.568627119064331,
708
+ "learning_rate": 0.00018231508365051922,
709
+ "loss": 1.313,
710
+ "step": 99
711
+ },
712
+ {
713
+ "epoch": 0.23,
714
+ "grad_norm": 0.4205207824707031,
715
+ "learning_rate": 0.0001818786520563467,
716
+ "loss": 1.3252,
717
+ "step": 100
718
+ },
719
+ {
720
+ "epoch": 0.23,
721
+ "grad_norm": 0.5135630369186401,
722
+ "learning_rate": 0.00018143743653376942,
723
+ "loss": 1.093,
724
+ "step": 101
725
+ },
726
+ {
727
+ "epoch": 0.24,
728
+ "grad_norm": 0.558722198009491,
729
+ "learning_rate": 0.0001809914628617105,
730
+ "loss": 1.5598,
731
+ "step": 102
732
+ },
733
+ {
734
+ "epoch": 0.24,
735
+ "grad_norm": 0.364212304353714,
736
+ "learning_rate": 0.00018054075709709756,
737
+ "loss": 1.1574,
738
+ "step": 103
739
+ },
740
+ {
741
+ "epoch": 0.24,
742
+ "grad_norm": 0.6438787579536438,
743
+ "learning_rate": 0.00018008534557334064,
744
+ "loss": 1.2912,
745
+ "step": 104
746
+ },
747
+ {
748
+ "epoch": 0.24,
749
+ "grad_norm": 0.4466971755027771,
750
+ "learning_rate": 0.00017962525489879325,
751
+ "loss": 1.2194,
752
+ "step": 105
753
+ },
754
+ {
755
+ "epoch": 0.25,
756
+ "grad_norm": 0.5050408244132996,
757
+ "learning_rate": 0.00017916051195519797,
758
+ "loss": 0.9488,
759
+ "step": 106
760
+ },
761
+ {
762
+ "epoch": 0.25,
763
+ "grad_norm": 0.4414922893047333,
764
+ "learning_rate": 0.00017869114389611575,
765
+ "loss": 0.9977,
766
+ "step": 107
767
+ },
768
+ {
769
+ "epoch": 0.25,
770
+ "grad_norm": 0.4354766011238098,
771
+ "learning_rate": 0.0001782171781453394,
772
+ "loss": 1.1738,
773
+ "step": 108
774
+ },
775
+ {
776
+ "epoch": 0.25,
777
+ "eval_loss": 1.1372472047805786,
778
+ "eval_runtime": 4.9728,
779
+ "eval_samples_per_second": 20.109,
780
+ "eval_steps_per_second": 20.109,
781
+ "step": 108
782
+ },
783
+ {
784
+ "epoch": 0.25,
785
+ "grad_norm": 0.44711774587631226,
786
+ "learning_rate": 0.00017773864239529132,
787
+ "loss": 1.0631,
788
+ "step": 109
789
+ },
790
+ {
791
+ "epoch": 0.26,
792
+ "grad_norm": 0.43031662702560425,
793
+ "learning_rate": 0.0001772555646054055,
794
+ "loss": 1.0922,
795
+ "step": 110
796
+ },
797
+ {
798
+ "epoch": 0.26,
799
+ "grad_norm": 0.4053135812282562,
800
+ "learning_rate": 0.00017676797300049393,
801
+ "loss": 1.2369,
802
+ "step": 111
803
+ },
804
+ {
805
+ "epoch": 0.26,
806
+ "grad_norm": 0.6041821241378784,
807
+ "learning_rate": 0.00017627589606909755,
808
+ "loss": 1.2172,
809
+ "step": 112
810
+ },
811
+ {
812
+ "epoch": 0.26,
813
+ "grad_norm": 0.2745780050754547,
814
+ "learning_rate": 0.00017577936256182167,
815
+ "loss": 0.4539,
816
+ "step": 113
817
+ },
818
+ {
819
+ "epoch": 0.26,
820
+ "grad_norm": 0.4951040744781494,
821
+ "learning_rate": 0.0001752784014896562,
822
+ "loss": 1.3143,
823
+ "step": 114
824
+ },
825
+ {
826
+ "epoch": 0.27,
827
+ "grad_norm": 0.38725313544273376,
828
+ "learning_rate": 0.00017477304212228057,
829
+ "loss": 1.0367,
830
+ "step": 115
831
+ },
832
+ {
833
+ "epoch": 0.27,
834
+ "grad_norm": 0.4349636435508728,
835
+ "learning_rate": 0.0001742633139863538,
836
+ "loss": 1.1372,
837
+ "step": 116
838
+ },
839
+ {
840
+ "epoch": 0.27,
841
+ "grad_norm": 0.5229642391204834,
842
+ "learning_rate": 0.00017374924686378905,
843
+ "loss": 1.2274,
844
+ "step": 117
845
+ },
846
+ {
847
+ "epoch": 0.27,
848
+ "grad_norm": 0.5884748101234436,
849
+ "learning_rate": 0.0001732308707900137,
850
+ "loss": 1.1546,
851
+ "step": 118
852
+ },
853
+ {
854
+ "epoch": 0.28,
855
+ "grad_norm": 0.4116392433643341,
856
+ "learning_rate": 0.0001727082160522145,
857
+ "loss": 1.1144,
858
+ "step": 119
859
+ },
860
+ {
861
+ "epoch": 0.28,
862
+ "grad_norm": 0.4359859228134155,
863
+ "learning_rate": 0.0001721813131875679,
864
+ "loss": 1.0402,
865
+ "step": 120
866
+ },
867
+ {
868
+ "epoch": 0.28,
869
+ "grad_norm": 0.4674035906791687,
870
+ "learning_rate": 0.00017165019298145585,
871
+ "loss": 0.7442,
872
+ "step": 121
873
+ },
874
+ {
875
+ "epoch": 0.28,
876
+ "grad_norm": 0.40642812848091125,
877
+ "learning_rate": 0.00017111488646566727,
878
+ "loss": 1.1118,
879
+ "step": 122
880
+ },
881
+ {
882
+ "epoch": 0.29,
883
+ "grad_norm": 0.4694182872772217,
884
+ "learning_rate": 0.00017057542491658468,
885
+ "loss": 1.3227,
886
+ "step": 123
887
+ },
888
+ {
889
+ "epoch": 0.29,
890
+ "grad_norm": 0.4917917251586914,
891
+ "learning_rate": 0.000170031839853357,
892
+ "loss": 1.3724,
893
+ "step": 124
894
+ },
895
+ {
896
+ "epoch": 0.29,
897
+ "grad_norm": 0.47446152567863464,
898
+ "learning_rate": 0.00016948416303605795,
899
+ "loss": 1.406,
900
+ "step": 125
901
+ },
902
+ {
903
+ "epoch": 0.29,
904
+ "grad_norm": 0.36553582549095154,
905
+ "learning_rate": 0.0001689324264638304,
906
+ "loss": 0.8186,
907
+ "step": 126
908
+ },
909
+ {
910
+ "epoch": 0.29,
911
+ "grad_norm": 0.3034989535808563,
912
+ "learning_rate": 0.00016837666237301663,
913
+ "loss": 0.5963,
914
+ "step": 127
915
+ },
916
+ {
917
+ "epoch": 0.3,
918
+ "grad_norm": 0.5044969320297241,
919
+ "learning_rate": 0.00016781690323527511,
920
+ "loss": 1.1356,
921
+ "step": 128
922
+ },
923
+ {
924
+ "epoch": 0.3,
925
+ "grad_norm": 0.449246346950531,
926
+ "learning_rate": 0.00016725318175568306,
927
+ "loss": 1.015,
928
+ "step": 129
929
+ },
930
+ {
931
+ "epoch": 0.3,
932
+ "grad_norm": 0.41430914402008057,
933
+ "learning_rate": 0.00016668553087082567,
934
+ "loss": 1.2896,
935
+ "step": 130
936
+ },
937
+ {
938
+ "epoch": 0.3,
939
+ "grad_norm": 0.4527166783809662,
940
+ "learning_rate": 0.0001661139837468717,
941
+ "loss": 1.3077,
942
+ "step": 131
943
+ },
944
+ {
945
+ "epoch": 0.31,
946
+ "grad_norm": 0.37175431847572327,
947
+ "learning_rate": 0.00016553857377763566,
948
+ "loss": 1.0345,
949
+ "step": 132
950
+ },
951
+ {
952
+ "epoch": 0.31,
953
+ "grad_norm": 0.5722200870513916,
954
+ "learning_rate": 0.0001649593345826268,
955
+ "loss": 1.2999,
956
+ "step": 133
957
+ },
958
+ {
959
+ "epoch": 0.31,
960
+ "grad_norm": 0.4093303978443146,
961
+ "learning_rate": 0.00016437630000508464,
962
+ "loss": 0.9219,
963
+ "step": 134
964
+ },
965
+ {
966
+ "epoch": 0.31,
967
+ "grad_norm": 0.5196316242218018,
968
+ "learning_rate": 0.00016378950411000183,
969
+ "loss": 1.2376,
970
+ "step": 135
971
+ },
972
+ {
973
+ "epoch": 0.32,
974
+ "grad_norm": 0.5437043905258179,
975
+ "learning_rate": 0.00016319898118213365,
976
+ "loss": 1.0182,
977
+ "step": 136
978
+ },
979
+ {
980
+ "epoch": 0.32,
981
+ "grad_norm": 0.5286529660224915,
982
+ "learning_rate": 0.00016260476572399496,
983
+ "loss": 1.4231,
984
+ "step": 137
985
+ },
986
+ {
987
+ "epoch": 0.32,
988
+ "grad_norm": 0.43187054991722107,
989
+ "learning_rate": 0.00016200689245384424,
990
+ "loss": 1.0627,
991
+ "step": 138
992
+ },
993
+ {
994
+ "epoch": 0.32,
995
+ "grad_norm": 0.6503031849861145,
996
+ "learning_rate": 0.00016140539630365522,
997
+ "loss": 1.1164,
998
+ "step": 139
999
+ },
1000
+ {
1001
+ "epoch": 0.32,
1002
+ "grad_norm": 0.3417191505432129,
1003
+ "learning_rate": 0.00016080031241707578,
1004
+ "loss": 0.9242,
1005
+ "step": 140
1006
+ },
1007
+ {
1008
+ "epoch": 0.33,
1009
+ "grad_norm": 0.4407976567745209,
1010
+ "learning_rate": 0.0001601916761473747,
1011
+ "loss": 1.1057,
1012
+ "step": 141
1013
+ },
1014
+ {
1015
+ "epoch": 0.33,
1016
+ "grad_norm": 0.5504916310310364,
1017
+ "learning_rate": 0.00015957952305537597,
1018
+ "loss": 1.1505,
1019
+ "step": 142
1020
+ },
1021
+ {
1022
+ "epoch": 0.33,
1023
+ "grad_norm": 0.44324642419815063,
1024
+ "learning_rate": 0.00015896388890738127,
1025
+ "loss": 1.2143,
1026
+ "step": 143
1027
+ },
1028
+ {
1029
+ "epoch": 0.33,
1030
+ "grad_norm": 0.44402655959129333,
1031
+ "learning_rate": 0.00015834480967308003,
1032
+ "loss": 1.0909,
1033
+ "step": 144
1034
+ },
1035
+ {
1036
+ "epoch": 0.34,
1037
+ "grad_norm": 0.539313554763794,
1038
+ "learning_rate": 0.00015772232152344795,
1039
+ "loss": 1.2159,
1040
+ "step": 145
1041
+ },
1042
+ {
1043
+ "epoch": 0.34,
1044
+ "grad_norm": 0.5022640824317932,
1045
+ "learning_rate": 0.0001570964608286336,
1046
+ "loss": 1.0763,
1047
+ "step": 146
1048
+ },
1049
+ {
1050
+ "epoch": 0.34,
1051
+ "grad_norm": 0.4640175402164459,
1052
+ "learning_rate": 0.00015646726415583344,
1053
+ "loss": 0.8441,
1054
+ "step": 147
1055
+ },
1056
+ {
1057
+ "epoch": 0.34,
1058
+ "grad_norm": 0.47599032521247864,
1059
+ "learning_rate": 0.0001558347682671553,
1060
+ "loss": 1.4246,
1061
+ "step": 148
1062
+ },
1063
+ {
1064
+ "epoch": 0.35,
1065
+ "grad_norm": 0.6268883943557739,
1066
+ "learning_rate": 0.00015519901011747044,
1067
+ "loss": 1.0982,
1068
+ "step": 149
1069
+ },
1070
+ {
1071
+ "epoch": 0.35,
1072
+ "grad_norm": 0.5821301937103271,
1073
+ "learning_rate": 0.00015456002685225448,
1074
+ "loss": 1.3526,
1075
+ "step": 150
1076
+ },
1077
+ {
1078
+ "epoch": 0.35,
1079
+ "grad_norm": 0.6036801934242249,
1080
+ "learning_rate": 0.00015391785580541698,
1081
+ "loss": 1.3432,
1082
+ "step": 151
1083
+ },
1084
+ {
1085
+ "epoch": 0.35,
1086
+ "grad_norm": 0.44567015767097473,
1087
+ "learning_rate": 0.0001532725344971202,
1088
+ "loss": 1.6177,
1089
+ "step": 152
1090
+ },
1091
+ {
1092
+ "epoch": 0.35,
1093
+ "grad_norm": 0.5531861782073975,
1094
+ "learning_rate": 0.0001526241006315869,
1095
+ "loss": 1.4035,
1096
+ "step": 153
1097
+ },
1098
+ {
1099
+ "epoch": 0.36,
1100
+ "grad_norm": 0.40534013509750366,
1101
+ "learning_rate": 0.00015197259209489747,
1102
+ "loss": 1.4051,
1103
+ "step": 154
1104
+ },
1105
+ {
1106
+ "epoch": 0.36,
1107
+ "grad_norm": 0.49140480160713196,
1108
+ "learning_rate": 0.00015131804695277612,
1109
+ "loss": 1.1617,
1110
+ "step": 155
1111
+ },
1112
+ {
1113
+ "epoch": 0.36,
1114
+ "grad_norm": 0.5026464462280273,
1115
+ "learning_rate": 0.00015066050344836706,
1116
+ "loss": 1.188,
1117
+ "step": 156
1118
+ },
1119
+ {
1120
+ "epoch": 0.36,
1121
+ "grad_norm": 0.454348623752594,
1122
+ "learning_rate": 0.00015000000000000001,
1123
+ "loss": 1.2269,
1124
+ "step": 157
1125
+ },
1126
+ {
1127
+ "epoch": 0.37,
1128
+ "grad_norm": 0.3233983516693115,
1129
+ "learning_rate": 0.0001493365751989454,
1130
+ "loss": 0.839,
1131
+ "step": 158
1132
+ },
1133
+ {
1134
+ "epoch": 0.37,
1135
+ "grad_norm": 0.35600027441978455,
1136
+ "learning_rate": 0.0001486702678071598,
1137
+ "loss": 0.8139,
1138
+ "step": 159
1139
+ },
1140
+ {
1141
+ "epoch": 0.37,
1142
+ "grad_norm": 0.3790452480316162,
1143
+ "learning_rate": 0.00014800111675502094,
1144
+ "loss": 0.9154,
1145
+ "step": 160
1146
+ },
1147
+ {
1148
+ "epoch": 0.37,
1149
+ "grad_norm": 0.3982885777950287,
1150
+ "learning_rate": 0.00014732916113905335,
1151
+ "loss": 1.1825,
1152
+ "step": 161
1153
+ },
1154
+ {
1155
+ "epoch": 0.38,
1156
+ "grad_norm": 0.5747079253196716,
1157
+ "learning_rate": 0.0001466544402196439,
1158
+ "loss": 1.0599,
1159
+ "step": 162
1160
+ },
1161
+ {
1162
+ "epoch": 0.38,
1163
+ "grad_norm": 0.412662148475647,
1164
+ "learning_rate": 0.00014597699341874806,
1165
+ "loss": 0.7421,
1166
+ "step": 163
1167
+ },
1168
+ {
1169
+ "epoch": 0.38,
1170
+ "grad_norm": 0.4285506308078766,
1171
+ "learning_rate": 0.00014529686031758643,
1172
+ "loss": 0.8738,
1173
+ "step": 164
1174
+ },
1175
+ {
1176
+ "epoch": 0.38,
1177
+ "grad_norm": 0.5278518199920654,
1178
+ "learning_rate": 0.00014461408065433227,
1179
+ "loss": 1.3702,
1180
+ "step": 165
1181
+ },
1182
+ {
1183
+ "epoch": 0.39,
1184
+ "grad_norm": 0.3949754536151886,
1185
+ "learning_rate": 0.00014392869432178971,
1186
+ "loss": 1.005,
1187
+ "step": 166
1188
+ },
1189
+ {
1190
+ "epoch": 0.39,
1191
+ "grad_norm": 0.39798328280448914,
1192
+ "learning_rate": 0.00014324074136506284,
1193
+ "loss": 1.1153,
1194
+ "step": 167
1195
+ },
1196
+ {
1197
+ "epoch": 0.39,
1198
+ "grad_norm": 0.6939488053321838,
1199
+ "learning_rate": 0.00014255026197921596,
1200
+ "loss": 1.0646,
1201
+ "step": 168
1202
+ },
1203
+ {
1204
+ "epoch": 0.39,
1205
+ "grad_norm": 0.45957663655281067,
1206
+ "learning_rate": 0.00014185729650692533,
1207
+ "loss": 1.2686,
1208
+ "step": 169
1209
+ },
1210
+ {
1211
+ "epoch": 0.39,
1212
+ "grad_norm": 0.4837738871574402,
1213
+ "learning_rate": 0.0001411618854361218,
1214
+ "loss": 1.3417,
1215
+ "step": 170
1216
+ },
1217
+ {
1218
+ "epoch": 0.4,
1219
+ "grad_norm": 0.6829615831375122,
1220
+ "learning_rate": 0.00014046406939762545,
1221
+ "loss": 1.1174,
1222
+ "step": 171
1223
+ },
1224
+ {
1225
+ "epoch": 0.4,
1226
+ "grad_norm": 0.5104063153266907,
1227
+ "learning_rate": 0.0001397638891627714,
1228
+ "loss": 1.2003,
1229
+ "step": 172
1230
+ },
1231
+ {
1232
+ "epoch": 0.4,
1233
+ "grad_norm": 0.6368303894996643,
1234
+ "learning_rate": 0.00013906138564102793,
1235
+ "loss": 1.2944,
1236
+ "step": 173
1237
+ },
1238
+ {
1239
+ "epoch": 0.4,
1240
+ "grad_norm": 0.5132100582122803,
1241
+ "learning_rate": 0.00013835659987760605,
1242
+ "loss": 1.3055,
1243
+ "step": 174
1244
+ },
1245
+ {
1246
+ "epoch": 0.41,
1247
+ "grad_norm": 0.4685085415840149,
1248
+ "learning_rate": 0.0001376495730510614,
1249
+ "loss": 1.3442,
1250
+ "step": 175
1251
+ },
1252
+ {
1253
+ "epoch": 0.41,
1254
+ "grad_norm": 0.5321069955825806,
1255
+ "learning_rate": 0.0001369403464708884,
1256
+ "loss": 1.2254,
1257
+ "step": 176
1258
+ },
1259
+ {
1260
+ "epoch": 0.41,
1261
+ "grad_norm": 0.5504405498504639,
1262
+ "learning_rate": 0.00013622896157510658,
1263
+ "loss": 1.306,
1264
+ "step": 177
1265
+ },
1266
+ {
1267
+ "epoch": 0.41,
1268
+ "grad_norm": 0.4058278203010559,
1269
+ "learning_rate": 0.00013551545992783947,
1270
+ "loss": 0.8795,
1271
+ "step": 178
1272
+ },
1273
+ {
1274
+ "epoch": 0.42,
1275
+ "grad_norm": 0.4485948979854584,
1276
+ "learning_rate": 0.0001347998832168862,
1277
+ "loss": 1.0097,
1278
+ "step": 179
1279
+ },
1280
+ {
1281
+ "epoch": 0.42,
1282
+ "grad_norm": 0.5311691164970398,
1283
+ "learning_rate": 0.0001340822732512857,
1284
+ "loss": 1.4547,
1285
+ "step": 180
1286
+ },
1287
+ {
1288
+ "epoch": 0.42,
1289
+ "grad_norm": 0.4577678442001343,
1290
+ "learning_rate": 0.00013336267195887398,
1291
+ "loss": 1.1929,
1292
+ "step": 181
1293
+ },
1294
+ {
1295
+ "epoch": 0.42,
1296
+ "grad_norm": 0.45450282096862793,
1297
+ "learning_rate": 0.00013264112138383445,
1298
+ "loss": 0.9774,
1299
+ "step": 182
1300
+ },
1301
+ {
1302
+ "epoch": 0.42,
1303
+ "grad_norm": 0.39516186714172363,
1304
+ "learning_rate": 0.00013191766368424133,
1305
+ "loss": 1.1093,
1306
+ "step": 183
1307
+ },
1308
+ {
1309
+ "epoch": 0.43,
1310
+ "grad_norm": 0.5446043014526367,
1311
+ "learning_rate": 0.00013119234112959655,
1312
+ "loss": 0.9583,
1313
+ "step": 184
1314
+ },
1315
+ {
1316
+ "epoch": 0.43,
1317
+ "grad_norm": 0.4918731451034546,
1318
+ "learning_rate": 0.00013046519609836,
1319
+ "loss": 1.35,
1320
+ "step": 185
1321
+ },
1322
+ {
1323
+ "epoch": 0.43,
1324
+ "grad_norm": 0.37299469113349915,
1325
+ "learning_rate": 0.00012973627107547346,
1326
+ "loss": 1.2582,
1327
+ "step": 186
1328
+ },
1329
+ {
1330
+ "epoch": 0.43,
1331
+ "grad_norm": 0.5013623237609863,
1332
+ "learning_rate": 0.0001290056086498785,
1333
+ "loss": 1.2556,
1334
+ "step": 187
1335
+ },
1336
+ {
1337
+ "epoch": 0.44,
1338
+ "grad_norm": 0.4400407373905182,
1339
+ "learning_rate": 0.00012827325151202782,
1340
+ "loss": 1.2137,
1341
+ "step": 188
1342
+ },
1343
+ {
1344
+ "epoch": 0.44,
1345
+ "grad_norm": 0.46342340111732483,
1346
+ "learning_rate": 0.00012753924245139135,
1347
+ "loss": 1.1706,
1348
+ "step": 189
1349
+ },
1350
+ {
1351
+ "epoch": 0.44,
1352
+ "grad_norm": 0.417937308549881,
1353
+ "learning_rate": 0.00012680362435395595,
1354
+ "loss": 1.1111,
1355
+ "step": 190
1356
+ },
1357
+ {
1358
+ "epoch": 0.44,
1359
+ "grad_norm": 0.9197636246681213,
1360
+ "learning_rate": 0.00012606644019971968,
1361
+ "loss": 1.4709,
1362
+ "step": 191
1363
+ },
1364
+ {
1365
+ "epoch": 0.45,
1366
+ "grad_norm": 0.4040437936782837,
1367
+ "learning_rate": 0.00012532773306018076,
1368
+ "loss": 1.1701,
1369
+ "step": 192
1370
+ },
1371
+ {
1372
+ "epoch": 0.45,
1373
+ "grad_norm": 0.5382991433143616,
1374
+ "learning_rate": 0.00012458754609582097,
1375
+ "loss": 0.9994,
1376
+ "step": 193
1377
+ },
1378
+ {
1379
+ "epoch": 0.45,
1380
+ "grad_norm": 0.40422263741493225,
1381
+ "learning_rate": 0.00012384592255358385,
1382
+ "loss": 1.0439,
1383
+ "step": 194
1384
+ },
1385
+ {
1386
+ "epoch": 0.45,
1387
+ "grad_norm": 0.5063545107841492,
1388
+ "learning_rate": 0.00012310290576434795,
1389
+ "loss": 1.4452,
1390
+ "step": 195
1391
+ },
1392
+ {
1393
+ "epoch": 0.45,
1394
+ "grad_norm": 0.37706631422042847,
1395
+ "learning_rate": 0.00012235853914039515,
1396
+ "loss": 0.9206,
1397
+ "step": 196
1398
+ },
1399
+ {
1400
+ "epoch": 0.46,
1401
+ "grad_norm": 0.4789075553417206,
1402
+ "learning_rate": 0.00012161286617287419,
1403
+ "loss": 0.9392,
1404
+ "step": 197
1405
+ },
1406
+ {
1407
+ "epoch": 0.46,
1408
+ "grad_norm": 0.43901360034942627,
1409
+ "learning_rate": 0.00012086593042925964,
1410
+ "loss": 1.228,
1411
+ "step": 198
1412
+ },
1413
+ {
1414
+ "epoch": 0.46,
1415
+ "grad_norm": 0.34066489338874817,
1416
+ "learning_rate": 0.00012011777555080638,
1417
+ "loss": 0.7,
1418
+ "step": 199
1419
+ },
1420
+ {
1421
+ "epoch": 0.46,
1422
+ "grad_norm": 0.5998871326446533,
1423
+ "learning_rate": 0.00011936844524999966,
1424
+ "loss": 1.3413,
1425
+ "step": 200
1426
+ },
1427
+ {
1428
+ "epoch": 0.47,
1429
+ "grad_norm": 0.43001553416252136,
1430
+ "learning_rate": 0.00011861798330800125,
1431
+ "loss": 1.0357,
1432
+ "step": 201
1433
+ },
1434
+ {
1435
+ "epoch": 0.47,
1436
+ "grad_norm": 0.5246945023536682,
1437
+ "learning_rate": 0.00011786643357209136,
1438
+ "loss": 0.9392,
1439
+ "step": 202
1440
+ },
1441
+ {
1442
+ "epoch": 0.47,
1443
+ "grad_norm": 0.614080548286438,
1444
+ "learning_rate": 0.00011711383995310681,
1445
+ "loss": 1.1147,
1446
+ "step": 203
1447
+ },
1448
+ {
1449
+ "epoch": 0.47,
1450
+ "grad_norm": 0.45178118348121643,
1451
+ "learning_rate": 0.00011636024642287546,
1452
+ "loss": 1.1284,
1453
+ "step": 204
1454
+ },
1455
+ {
1456
+ "epoch": 0.48,
1457
+ "grad_norm": 0.4508957266807556,
1458
+ "learning_rate": 0.00011560569701164697,
1459
+ "loss": 1.3169,
1460
+ "step": 205
1461
+ },
1462
+ {
1463
+ "epoch": 0.48,
1464
+ "grad_norm": 1.084119439125061,
1465
+ "learning_rate": 0.00011485023580552039,
1466
+ "loss": 1.4725,
1467
+ "step": 206
1468
+ },
1469
+ {
1470
+ "epoch": 0.48,
1471
+ "grad_norm": 0.417682409286499,
1472
+ "learning_rate": 0.00011409390694386817,
1473
+ "loss": 1.1748,
1474
+ "step": 207
1475
+ },
1476
+ {
1477
+ "epoch": 0.48,
1478
+ "grad_norm": 0.4242180585861206,
1479
+ "learning_rate": 0.00011333675461675739,
1480
+ "loss": 1.0729,
1481
+ "step": 208
1482
+ },
1483
+ {
1484
+ "epoch": 0.48,
1485
+ "grad_norm": 0.4738271236419678,
1486
+ "learning_rate": 0.00011257882306236775,
1487
+ "loss": 1.2974,
1488
+ "step": 209
1489
+ },
1490
+ {
1491
+ "epoch": 0.49,
1492
+ "grad_norm": 0.3936362862586975,
1493
+ "learning_rate": 0.00011182015656440692,
1494
+ "loss": 1.0202,
1495
+ "step": 210
1496
+ },
1497
+ {
1498
+ "epoch": 0.49,
1499
+ "grad_norm": 0.4622405767440796,
1500
+ "learning_rate": 0.00011106079944952317,
1501
+ "loss": 1.2899,
1502
+ "step": 211
1503
+ },
1504
+ {
1505
+ "epoch": 0.49,
1506
+ "grad_norm": 0.5150789618492126,
1507
+ "learning_rate": 0.00011030079608471544,
1508
+ "loss": 0.8933,
1509
+ "step": 212
1510
+ },
1511
+ {
1512
+ "epoch": 0.49,
1513
+ "grad_norm": 0.5435701608657837,
1514
+ "learning_rate": 0.00010954019087474124,
1515
+ "loss": 1.4871,
1516
+ "step": 213
1517
+ },
1518
+ {
1519
+ "epoch": 0.5,
1520
+ "grad_norm": 0.42681682109832764,
1521
+ "learning_rate": 0.00010877902825952197,
1522
+ "loss": 1.0943,
1523
+ "step": 214
1524
+ },
1525
+ {
1526
+ "epoch": 0.5,
1527
+ "grad_norm": 0.46130529046058655,
1528
+ "learning_rate": 0.00010801735271154669,
1529
+ "loss": 1.1492,
1530
+ "step": 215
1531
+ },
1532
+ {
1533
+ "epoch": 0.5,
1534
+ "grad_norm": 0.502712607383728,
1535
+ "learning_rate": 0.00010725520873327361,
1536
+ "loss": 1.1175,
1537
+ "step": 216
1538
+ },
1539
+ {
1540
+ "epoch": 0.5,
1541
+ "eval_loss": 1.1232960224151611,
1542
+ "eval_runtime": 4.7032,
1543
+ "eval_samples_per_second": 21.262,
1544
+ "eval_steps_per_second": 21.262,
1545
+ "step": 216
1546
+ },
1547
+ {
1548
+ "epoch": 0.5,
1549
+ "grad_norm": 0.4459720849990845,
1550
+ "learning_rate": 0.00010649264085452988,
1551
+ "loss": 0.9861,
1552
+ "step": 217
1553
+ },
1554
+ {
1555
+ "epoch": 0.51,
1556
+ "grad_norm": 0.5748346447944641,
1557
+ "learning_rate": 0.00010572969362990998,
1558
+ "loss": 1.2841,
1559
+ "step": 218
1560
+ },
1561
+ {
1562
+ "epoch": 0.51,
1563
+ "grad_norm": 0.5053660273551941,
1564
+ "learning_rate": 0.0001049664116361724,
1565
+ "loss": 1.1955,
1566
+ "step": 219
1567
+ },
1568
+ {
1569
+ "epoch": 0.51,
1570
+ "grad_norm": 0.5145143270492554,
1571
+ "learning_rate": 0.0001042028394696352,
1572
+ "loss": 1.0322,
1573
+ "step": 220
1574
+ },
1575
+ {
1576
+ "epoch": 0.51,
1577
+ "grad_norm": 0.552711009979248,
1578
+ "learning_rate": 0.00010343902174357039,
1579
+ "loss": 1.1417,
1580
+ "step": 221
1581
+ },
1582
+ {
1583
+ "epoch": 0.52,
1584
+ "grad_norm": 0.4026980400085449,
1585
+ "learning_rate": 0.00010267500308559732,
1586
+ "loss": 1.1334,
1587
+ "step": 222
1588
+ },
1589
+ {
1590
+ "epoch": 0.52,
1591
+ "grad_norm": 0.5193754434585571,
1592
+ "learning_rate": 0.0001019108281350752,
1593
+ "loss": 1.0735,
1594
+ "step": 223
1595
+ },
1596
+ {
1597
+ "epoch": 0.52,
1598
+ "grad_norm": 0.4189368486404419,
1599
+ "learning_rate": 0.0001011465415404949,
1600
+ "loss": 1.2567,
1601
+ "step": 224
1602
+ },
1603
+ {
1604
+ "epoch": 0.52,
1605
+ "grad_norm": 0.45475542545318604,
1606
+ "learning_rate": 0.0001003821879568704,
1607
+ "loss": 0.9511,
1608
+ "step": 225
1609
+ },
1610
+ {
1611
+ "epoch": 0.52,
1612
+ "grad_norm": 0.46022626757621765,
1613
+ "learning_rate": 9.96178120431296e-05,
1614
+ "loss": 1.1802,
1615
+ "step": 226
1616
+ },
1617
+ {
1618
+ "epoch": 0.53,
1619
+ "grad_norm": 0.4458233416080475,
1620
+ "learning_rate": 9.88534584595051e-05,
1621
+ "loss": 1.0332,
1622
+ "step": 227
1623
+ },
1624
+ {
1625
+ "epoch": 0.53,
1626
+ "grad_norm": 0.3980939984321594,
1627
+ "learning_rate": 9.80891718649248e-05,
1628
+ "loss": 0.9659,
1629
+ "step": 228
1630
+ },
1631
+ {
1632
+ "epoch": 0.53,
1633
+ "grad_norm": 0.4233875274658203,
1634
+ "learning_rate": 9.732499691440266e-05,
1635
+ "loss": 1.3747,
1636
+ "step": 229
1637
+ },
1638
+ {
1639
+ "epoch": 0.53,
1640
+ "grad_norm": 0.4073673486709595,
1641
+ "learning_rate": 9.656097825642961e-05,
1642
+ "loss": 1.2297,
1643
+ "step": 230
1644
+ },
1645
+ {
1646
+ "epoch": 0.54,
1647
+ "grad_norm": 0.39706096053123474,
1648
+ "learning_rate": 9.579716053036479e-05,
1649
+ "loss": 0.973,
1650
+ "step": 231
1651
+ },
1652
+ {
1653
+ "epoch": 0.54,
1654
+ "grad_norm": 0.4343310296535492,
1655
+ "learning_rate": 9.503358836382761e-05,
1656
+ "loss": 1.238,
1657
+ "step": 232
1658
+ },
1659
+ {
1660
+ "epoch": 0.54,
1661
+ "grad_norm": 0.4828486144542694,
1662
+ "learning_rate": 9.427030637009003e-05,
1663
+ "loss": 1.1616,
1664
+ "step": 233
1665
+ },
1666
+ {
1667
+ "epoch": 0.54,
1668
+ "grad_norm": 0.56890869140625,
1669
+ "learning_rate": 9.35073591454701e-05,
1670
+ "loss": 0.8345,
1671
+ "step": 234
1672
+ },
1673
+ {
1674
+ "epoch": 0.55,
1675
+ "grad_norm": 0.4366269111633301,
1676
+ "learning_rate": 9.274479126672641e-05,
1677
+ "loss": 1.111,
1678
+ "step": 235
1679
+ },
1680
+ {
1681
+ "epoch": 0.55,
1682
+ "grad_norm": 0.4700612723827362,
1683
+ "learning_rate": 9.198264728845332e-05,
1684
+ "loss": 1.3553,
1685
+ "step": 236
1686
+ },
1687
+ {
1688
+ "epoch": 0.55,
1689
+ "grad_norm": 0.40685421228408813,
1690
+ "learning_rate": 9.122097174047805e-05,
1691
+ "loss": 1.1515,
1692
+ "step": 237
1693
+ },
1694
+ {
1695
+ "epoch": 0.55,
1696
+ "grad_norm": 0.37611886858940125,
1697
+ "learning_rate": 9.045980912525879e-05,
1698
+ "loss": 0.9226,
1699
+ "step": 238
1700
+ },
1701
+ {
1702
+ "epoch": 0.55,
1703
+ "grad_norm": 0.40409284830093384,
1704
+ "learning_rate": 8.969920391528458e-05,
1705
+ "loss": 0.8641,
1706
+ "step": 239
1707
+ },
1708
+ {
1709
+ "epoch": 0.56,
1710
+ "grad_norm": 0.39749234914779663,
1711
+ "learning_rate": 8.893920055047686e-05,
1712
+ "loss": 1.0498,
1713
+ "step": 240
1714
+ },
1715
+ {
1716
+ "epoch": 0.56,
1717
+ "grad_norm": 0.554320752620697,
1718
+ "learning_rate": 8.81798434355931e-05,
1719
+ "loss": 1.1806,
1720
+ "step": 241
1721
+ },
1722
+ {
1723
+ "epoch": 0.56,
1724
+ "grad_norm": 0.37640953063964844,
1725
+ "learning_rate": 8.742117693763227e-05,
1726
+ "loss": 1.03,
1727
+ "step": 242
1728
+ },
1729
+ {
1730
+ "epoch": 0.56,
1731
+ "grad_norm": 0.4188061058521271,
1732
+ "learning_rate": 8.666324538324264e-05,
1733
+ "loss": 1.0891,
1734
+ "step": 243
1735
+ },
1736
+ {
1737
+ "epoch": 0.57,
1738
+ "grad_norm": 0.40569260716438293,
1739
+ "learning_rate": 8.590609305613184e-05,
1740
+ "loss": 1.2237,
1741
+ "step": 244
1742
+ },
1743
+ {
1744
+ "epoch": 0.57,
1745
+ "grad_norm": 0.35996559262275696,
1746
+ "learning_rate": 8.514976419447964e-05,
1747
+ "loss": 1.1086,
1748
+ "step": 245
1749
+ },
1750
+ {
1751
+ "epoch": 0.57,
1752
+ "grad_norm": 0.3959590792655945,
1753
+ "learning_rate": 8.439430298835304e-05,
1754
+ "loss": 1.2445,
1755
+ "step": 246
1756
+ },
1757
+ {
1758
+ "epoch": 0.57,
1759
+ "grad_norm": 0.41735970973968506,
1760
+ "learning_rate": 8.363975357712457e-05,
1761
+ "loss": 1.2318,
1762
+ "step": 247
1763
+ },
1764
+ {
1765
+ "epoch": 0.58,
1766
+ "grad_norm": 0.40587231516838074,
1767
+ "learning_rate": 8.28861600468932e-05,
1768
+ "loss": 0.8108,
1769
+ "step": 248
1770
+ },
1771
+ {
1772
+ "epoch": 0.58,
1773
+ "grad_norm": 0.4381011128425598,
1774
+ "learning_rate": 8.213356642790867e-05,
1775
+ "loss": 0.8791,
1776
+ "step": 249
1777
+ },
1778
+ {
1779
+ "epoch": 0.58,
1780
+ "grad_norm": 0.455411821603775,
1781
+ "learning_rate": 8.138201669199879e-05,
1782
+ "loss": 1.1067,
1783
+ "step": 250
1784
+ },
1785
+ {
1786
+ "epoch": 0.58,
1787
+ "grad_norm": 0.486929714679718,
1788
+ "learning_rate": 8.063155475000037e-05,
1789
+ "loss": 1.0332,
1790
+ "step": 251
1791
+ },
1792
+ {
1793
+ "epoch": 0.58,
1794
+ "grad_norm": 0.3872472941875458,
1795
+ "learning_rate": 7.988222444919364e-05,
1796
+ "loss": 0.9779,
1797
+ "step": 252
1798
+ },
1799
+ {
1800
+ "epoch": 0.59,
1801
+ "grad_norm": 0.5021295547485352,
1802
+ "learning_rate": 7.913406957074037e-05,
1803
+ "loss": 1.2159,
1804
+ "step": 253
1805
+ },
1806
+ {
1807
+ "epoch": 0.59,
1808
+ "grad_norm": 0.3245134949684143,
1809
+ "learning_rate": 7.838713382712583e-05,
1810
+ "loss": 0.6035,
1811
+ "step": 254
1812
+ },
1813
+ {
1814
+ "epoch": 0.59,
1815
+ "grad_norm": 0.39470720291137695,
1816
+ "learning_rate": 7.76414608596049e-05,
1817
+ "loss": 1.0058,
1818
+ "step": 255
1819
+ },
1820
+ {
1821
+ "epoch": 0.59,
1822
+ "grad_norm": 0.36348241567611694,
1823
+ "learning_rate": 7.68970942356521e-05,
1824
+ "loss": 0.6888,
1825
+ "step": 256
1826
+ },
1827
+ {
1828
+ "epoch": 0.6,
1829
+ "grad_norm": 0.6455038189888,
1830
+ "learning_rate": 7.615407744641619e-05,
1831
+ "loss": 1.0407,
1832
+ "step": 257
1833
+ },
1834
+ {
1835
+ "epoch": 0.6,
1836
+ "grad_norm": 0.5500055551528931,
1837
+ "learning_rate": 7.541245390417906e-05,
1838
+ "loss": 1.1612,
1839
+ "step": 258
1840
+ },
1841
+ {
1842
+ "epoch": 0.6,
1843
+ "grad_norm": 0.3761384189128876,
1844
+ "learning_rate": 7.467226693981925e-05,
1845
+ "loss": 0.9731,
1846
+ "step": 259
1847
+ },
1848
+ {
1849
+ "epoch": 0.6,
1850
+ "grad_norm": 0.6317098736763,
1851
+ "learning_rate": 7.393355980028039e-05,
1852
+ "loss": 1.3037,
1853
+ "step": 260
1854
+ },
1855
+ {
1856
+ "epoch": 0.61,
1857
+ "grad_norm": 0.4043276607990265,
1858
+ "learning_rate": 7.319637564604412e-05,
1859
+ "loss": 1.323,
1860
+ "step": 261
1861
+ },
1862
+ {
1863
+ "epoch": 0.61,
1864
+ "grad_norm": 0.46140003204345703,
1865
+ "learning_rate": 7.246075754860868e-05,
1866
+ "loss": 0.8617,
1867
+ "step": 262
1868
+ },
1869
+ {
1870
+ "epoch": 0.61,
1871
+ "grad_norm": 0.4819968044757843,
1872
+ "learning_rate": 7.172674848797219e-05,
1873
+ "loss": 1.1345,
1874
+ "step": 263
1875
+ },
1876
+ {
1877
+ "epoch": 0.61,
1878
+ "grad_norm": 0.46252205967903137,
1879
+ "learning_rate": 7.099439135012153e-05,
1880
+ "loss": 1.4092,
1881
+ "step": 264
1882
+ },
1883
+ {
1884
+ "epoch": 0.61,
1885
+ "grad_norm": 0.5231882333755493,
1886
+ "learning_rate": 7.026372892452653e-05,
1887
+ "loss": 1.1937,
1888
+ "step": 265
1889
+ },
1890
+ {
1891
+ "epoch": 0.62,
1892
+ "grad_norm": 0.3481261432170868,
1893
+ "learning_rate": 6.953480390164e-05,
1894
+ "loss": 0.7882,
1895
+ "step": 266
1896
+ },
1897
+ {
1898
+ "epoch": 0.62,
1899
+ "grad_norm": 0.8361053466796875,
1900
+ "learning_rate": 6.880765887040343e-05,
1901
+ "loss": 1.4242,
1902
+ "step": 267
1903
+ },
1904
+ {
1905
+ "epoch": 0.62,
1906
+ "grad_norm": 0.3501080572605133,
1907
+ "learning_rate": 6.808233631575867e-05,
1908
+ "loss": 0.7428,
1909
+ "step": 268
1910
+ },
1911
+ {
1912
+ "epoch": 0.62,
1913
+ "grad_norm": 0.46989908814430237,
1914
+ "learning_rate": 6.735887861616556e-05,
1915
+ "loss": 1.2293,
1916
+ "step": 269
1917
+ },
1918
+ {
1919
+ "epoch": 0.63,
1920
+ "grad_norm": 0.44108134508132935,
1921
+ "learning_rate": 6.663732804112603e-05,
1922
+ "loss": 1.1302,
1923
+ "step": 270
1924
+ },
1925
+ {
1926
+ "epoch": 0.63,
1927
+ "grad_norm": 0.484967440366745,
1928
+ "learning_rate": 6.591772674871434e-05,
1929
+ "loss": 1.2551,
1930
+ "step": 271
1931
+ },
1932
+ {
1933
+ "epoch": 0.63,
1934
+ "grad_norm": 0.4039287865161896,
1935
+ "learning_rate": 6.520011678311382e-05,
1936
+ "loss": 1.2437,
1937
+ "step": 272
1938
+ },
1939
+ {
1940
+ "epoch": 0.63,
1941
+ "grad_norm": 0.5447697639465332,
1942
+ "learning_rate": 6.448454007216054e-05,
1943
+ "loss": 1.1141,
1944
+ "step": 273
1945
+ },
1946
+ {
1947
+ "epoch": 0.64,
1948
+ "grad_norm": 0.5546038150787354,
1949
+ "learning_rate": 6.377103842489343e-05,
1950
+ "loss": 0.8614,
1951
+ "step": 274
1952
+ },
1953
+ {
1954
+ "epoch": 0.64,
1955
+ "grad_norm": 0.4008563756942749,
1956
+ "learning_rate": 6.305965352911161e-05,
1957
+ "loss": 0.9837,
1958
+ "step": 275
1959
+ },
1960
+ {
1961
+ "epoch": 0.64,
1962
+ "grad_norm": 0.4528581202030182,
1963
+ "learning_rate": 6.235042694893862e-05,
1964
+ "loss": 1.3003,
1965
+ "step": 276
1966
+ },
1967
+ {
1968
+ "epoch": 0.64,
1969
+ "grad_norm": 0.5693272352218628,
1970
+ "learning_rate": 6.164340012239396e-05,
1971
+ "loss": 1.2459,
1972
+ "step": 277
1973
+ },
1974
+ {
1975
+ "epoch": 0.65,
1976
+ "grad_norm": 0.44974976778030396,
1977
+ "learning_rate": 6.093861435897208e-05,
1978
+ "loss": 1.3594,
1979
+ "step": 278
1980
+ },
1981
+ {
1982
+ "epoch": 0.65,
1983
+ "grad_norm": 0.42815354466438293,
1984
+ "learning_rate": 6.02361108372286e-05,
1985
+ "loss": 1.1973,
1986
+ "step": 279
1987
+ },
1988
+ {
1989
+ "epoch": 0.65,
1990
+ "grad_norm": 0.4000280797481537,
1991
+ "learning_rate": 5.953593060237457e-05,
1992
+ "loss": 1.1897,
1993
+ "step": 280
1994
+ },
1995
+ {
1996
+ "epoch": 0.65,
1997
+ "grad_norm": 0.4146158993244171,
1998
+ "learning_rate": 5.883811456387821e-05,
1999
+ "loss": 0.9858,
2000
+ "step": 281
2001
+ },
2002
+ {
2003
+ "epoch": 0.65,
2004
+ "grad_norm": 0.43319183588027954,
2005
+ "learning_rate": 5.8142703493074714e-05,
2006
+ "loss": 0.9849,
2007
+ "step": 282
2008
+ },
2009
+ {
2010
+ "epoch": 0.66,
2011
+ "grad_norm": 0.4292287528514862,
2012
+ "learning_rate": 5.7449738020784085e-05,
2013
+ "loss": 1.2853,
2014
+ "step": 283
2015
+ },
2016
+ {
2017
+ "epoch": 0.66,
2018
+ "grad_norm": 0.4380340576171875,
2019
+ "learning_rate": 5.675925863493721e-05,
2020
+ "loss": 1.1053,
2021
+ "step": 284
2022
+ },
2023
+ {
2024
+ "epoch": 0.66,
2025
+ "grad_norm": 0.47855687141418457,
2026
+ "learning_rate": 5.607130567821031e-05,
2027
+ "loss": 1.147,
2028
+ "step": 285
2029
+ },
2030
+ {
2031
+ "epoch": 0.66,
2032
+ "grad_norm": 0.47737643122673035,
2033
+ "learning_rate": 5.5385919345667715e-05,
2034
+ "loss": 1.2292,
2035
+ "step": 286
2036
+ },
2037
+ {
2038
+ "epoch": 0.67,
2039
+ "grad_norm": 0.532492995262146,
2040
+ "learning_rate": 5.4703139682413586e-05,
2041
+ "loss": 1.2948,
2042
+ "step": 287
2043
+ },
2044
+ {
2045
+ "epoch": 0.67,
2046
+ "grad_norm": 0.47192513942718506,
2047
+ "learning_rate": 5.402300658125197e-05,
2048
+ "loss": 1.0005,
2049
+ "step": 288
2050
+ },
2051
+ {
2052
+ "epoch": 0.67,
2053
+ "grad_norm": 0.3757496476173401,
2054
+ "learning_rate": 5.334555978035609e-05,
2055
+ "loss": 0.904,
2056
+ "step": 289
2057
+ },
2058
+ {
2059
+ "epoch": 0.67,
2060
+ "grad_norm": 0.5040110349655151,
2061
+ "learning_rate": 5.267083886094668e-05,
2062
+ "loss": 1.4786,
2063
+ "step": 290
2064
+ },
2065
+ {
2066
+ "epoch": 0.68,
2067
+ "grad_norm": 0.4644010663032532,
2068
+ "learning_rate": 5.199888324497907e-05,
2069
+ "loss": 1.0901,
2070
+ "step": 291
2071
+ },
2072
+ {
2073
+ "epoch": 0.68,
2074
+ "grad_norm": 0.4540674686431885,
2075
+ "learning_rate": 5.132973219284023e-05,
2076
+ "loss": 1.1761,
2077
+ "step": 292
2078
+ },
2079
+ {
2080
+ "epoch": 0.68,
2081
+ "grad_norm": 0.4109453558921814,
2082
+ "learning_rate": 5.0663424801054595e-05,
2083
+ "loss": 1.1454,
2084
+ "step": 293
2085
+ },
2086
+ {
2087
+ "epoch": 0.68,
2088
+ "grad_norm": 0.46687519550323486,
2089
+ "learning_rate": 5.000000000000002e-05,
2090
+ "loss": 1.0732,
2091
+ "step": 294
2092
+ },
2093
+ {
2094
+ "epoch": 0.68,
2095
+ "grad_norm": 0.5146364569664001,
2096
+ "learning_rate": 4.9339496551632944e-05,
2097
+ "loss": 1.45,
2098
+ "step": 295
2099
+ },
2100
+ {
2101
+ "epoch": 0.69,
2102
+ "grad_norm": 0.4732467532157898,
2103
+ "learning_rate": 4.8681953047223914e-05,
2104
+ "loss": 0.8685,
2105
+ "step": 296
2106
+ },
2107
+ {
2108
+ "epoch": 0.69,
2109
+ "grad_norm": 0.5239700675010681,
2110
+ "learning_rate": 4.8027407905102585e-05,
2111
+ "loss": 1.1003,
2112
+ "step": 297
2113
+ },
2114
+ {
2115
+ "epoch": 0.69,
2116
+ "grad_norm": 0.404216468334198,
2117
+ "learning_rate": 4.73758993684131e-05,
2118
+ "loss": 0.8534,
2119
+ "step": 298
2120
+ },
2121
+ {
2122
+ "epoch": 0.69,
2123
+ "grad_norm": 0.3570495545864105,
2124
+ "learning_rate": 4.672746550287985e-05,
2125
+ "loss": 0.9244,
2126
+ "step": 299
2127
+ },
2128
+ {
2129
+ "epoch": 0.7,
2130
+ "grad_norm": 0.535670816898346,
2131
+ "learning_rate": 4.6082144194583056e-05,
2132
+ "loss": 1.4359,
2133
+ "step": 300
2134
+ },
2135
+ {
2136
+ "epoch": 0.7,
2137
+ "grad_norm": 0.4560551047325134,
2138
+ "learning_rate": 4.543997314774553e-05,
2139
+ "loss": 1.0442,
2140
+ "step": 301
2141
+ },
2142
+ {
2143
+ "epoch": 0.7,
2144
+ "grad_norm": 0.4430754482746124,
2145
+ "learning_rate": 4.4800989882529574e-05,
2146
+ "loss": 1.1429,
2147
+ "step": 302
2148
+ },
2149
+ {
2150
+ "epoch": 0.7,
2151
+ "grad_norm": 0.48371657729148865,
2152
+ "learning_rate": 4.41652317328447e-05,
2153
+ "loss": 1.21,
2154
+ "step": 303
2155
+ },
2156
+ {
2157
+ "epoch": 0.71,
2158
+ "grad_norm": 0.462596595287323,
2159
+ "learning_rate": 4.3532735844166574e-05,
2160
+ "loss": 1.0128,
2161
+ "step": 304
2162
+ },
2163
+ {
2164
+ "epoch": 0.71,
2165
+ "grad_norm": 0.43516165018081665,
2166
+ "learning_rate": 4.2903539171366393e-05,
2167
+ "loss": 1.1847,
2168
+ "step": 305
2169
+ },
2170
+ {
2171
+ "epoch": 0.71,
2172
+ "grad_norm": 0.805388867855072,
2173
+ "learning_rate": 4.227767847655205e-05,
2174
+ "loss": 1.2468,
2175
+ "step": 306
2176
+ },
2177
+ {
2178
+ "epoch": 0.71,
2179
+ "grad_norm": 0.4382609724998474,
2180
+ "learning_rate": 4.165519032691998e-05,
2181
+ "loss": 1.2178,
2182
+ "step": 307
2183
+ },
2184
+ {
2185
+ "epoch": 0.71,
2186
+ "grad_norm": 0.3396044969558716,
2187
+ "learning_rate": 4.1036111092618725e-05,
2188
+ "loss": 0.9241,
2189
+ "step": 308
2190
+ },
2191
+ {
2192
+ "epoch": 0.72,
2193
+ "grad_norm": 0.4570239782333374,
2194
+ "learning_rate": 4.042047694462404e-05,
2195
+ "loss": 1.0592,
2196
+ "step": 309
2197
+ },
2198
+ {
2199
+ "epoch": 0.72,
2200
+ "grad_norm": 0.49074164032936096,
2201
+ "learning_rate": 3.9808323852625316e-05,
2202
+ "loss": 1.1067,
2203
+ "step": 310
2204
+ },
2205
+ {
2206
+ "epoch": 0.72,
2207
+ "grad_norm": 0.5517829656600952,
2208
+ "learning_rate": 3.919968758292425e-05,
2209
+ "loss": 1.3361,
2210
+ "step": 311
2211
+ },
2212
+ {
2213
+ "epoch": 0.72,
2214
+ "grad_norm": 0.4893704950809479,
2215
+ "learning_rate": 3.859460369634479e-05,
2216
+ "loss": 1.1157,
2217
+ "step": 312
2218
+ },
2219
+ {
2220
+ "epoch": 0.73,
2221
+ "grad_norm": 0.4687519967556,
2222
+ "learning_rate": 3.799310754615578e-05,
2223
+ "loss": 1.3605,
2224
+ "step": 313
2225
+ },
2226
+ {
2227
+ "epoch": 0.73,
2228
+ "grad_norm": 0.35046279430389404,
2229
+ "learning_rate": 3.7395234276005087e-05,
2230
+ "loss": 0.8558,
2231
+ "step": 314
2232
+ },
2233
+ {
2234
+ "epoch": 0.73,
2235
+ "grad_norm": 0.5013456344604492,
2236
+ "learning_rate": 3.6801018817866375e-05,
2237
+ "loss": 1.0554,
2238
+ "step": 315
2239
+ },
2240
+ {
2241
+ "epoch": 0.73,
2242
+ "grad_norm": 0.4760110378265381,
2243
+ "learning_rate": 3.62104958899982e-05,
2244
+ "loss": 1.1839,
2245
+ "step": 316
2246
+ },
2247
+ {
2248
+ "epoch": 0.74,
2249
+ "grad_norm": 0.5143394470214844,
2250
+ "learning_rate": 3.562369999491536e-05,
2251
+ "loss": 1.4146,
2252
+ "step": 317
2253
+ },
2254
+ {
2255
+ "epoch": 0.74,
2256
+ "grad_norm": 0.4014797508716583,
2257
+ "learning_rate": 3.504066541737323e-05,
2258
+ "loss": 1.0731,
2259
+ "step": 318
2260
+ },
2261
+ {
2262
+ "epoch": 0.74,
2263
+ "grad_norm": 0.40402811765670776,
2264
+ "learning_rate": 3.4461426222364336e-05,
2265
+ "loss": 1.0001,
2266
+ "step": 319
2267
+ },
2268
+ {
2269
+ "epoch": 0.74,
2270
+ "grad_norm": 0.43316271901130676,
2271
+ "learning_rate": 3.3886016253128326e-05,
2272
+ "loss": 1.1141,
2273
+ "step": 320
2274
+ },
2275
+ {
2276
+ "epoch": 0.74,
2277
+ "grad_norm": 0.49771907925605774,
2278
+ "learning_rate": 3.3314469129174364e-05,
2279
+ "loss": 1.3224,
2280
+ "step": 321
2281
+ },
2282
+ {
2283
+ "epoch": 0.75,
2284
+ "grad_norm": 0.4895000159740448,
2285
+ "learning_rate": 3.2746818244316956e-05,
2286
+ "loss": 1.04,
2287
+ "step": 322
2288
+ },
2289
+ {
2290
+ "epoch": 0.75,
2291
+ "grad_norm": 0.5450279116630554,
2292
+ "learning_rate": 3.2183096764724915e-05,
2293
+ "loss": 1.4016,
2294
+ "step": 323
2295
+ },
2296
+ {
2297
+ "epoch": 0.75,
2298
+ "grad_norm": 0.4866601526737213,
2299
+ "learning_rate": 3.16233376269834e-05,
2300
+ "loss": 1.4102,
2301
+ "step": 324
2302
+ },
2303
+ {
2304
+ "epoch": 0.75,
2305
+ "eval_loss": 1.1097332239151,
2306
+ "eval_runtime": 5.825,
2307
+ "eval_samples_per_second": 17.167,
2308
+ "eval_steps_per_second": 17.167,
2309
+ "step": 324
2310
+ },
2311
+ {
2312
+ "epoch": 0.75,
2313
+ "grad_norm": 0.40248391032218933,
2314
+ "learning_rate": 3.106757353616966e-05,
2315
+ "loss": 0.8404,
2316
+ "step": 325
2317
+ },
2318
+ {
2319
+ "epoch": 0.76,
2320
+ "grad_norm": 0.44263288378715515,
2321
+ "learning_rate": 3.0515836963942056e-05,
2322
+ "loss": 1.2992,
2323
+ "step": 326
2324
+ },
2325
+ {
2326
+ "epoch": 0.76,
2327
+ "grad_norm": 0.4113296866416931,
2328
+ "learning_rate": 2.9968160146643022e-05,
2329
+ "loss": 1.1058,
2330
+ "step": 327
2331
+ },
2332
+ {
2333
+ "epoch": 0.76,
2334
+ "grad_norm": 0.5293351411819458,
2335
+ "learning_rate": 2.9424575083415362e-05,
2336
+ "loss": 1.5505,
2337
+ "step": 328
2338
+ },
2339
+ {
2340
+ "epoch": 0.76,
2341
+ "grad_norm": 0.500862181186676,
2342
+ "learning_rate": 2.888511353433274e-05,
2343
+ "loss": 1.1173,
2344
+ "step": 329
2345
+ },
2346
+ {
2347
+ "epoch": 0.77,
2348
+ "grad_norm": 0.408054918050766,
2349
+ "learning_rate": 2.8349807018544174e-05,
2350
+ "loss": 1.1951,
2351
+ "step": 330
2352
+ },
2353
+ {
2354
+ "epoch": 0.77,
2355
+ "grad_norm": 0.39368724822998047,
2356
+ "learning_rate": 2.7818686812432136e-05,
2357
+ "loss": 1.1842,
2358
+ "step": 331
2359
+ },
2360
+ {
2361
+ "epoch": 0.77,
2362
+ "grad_norm": 0.5162099599838257,
2363
+ "learning_rate": 2.7291783947785543e-05,
2364
+ "loss": 1.5018,
2365
+ "step": 332
2366
+ },
2367
+ {
2368
+ "epoch": 0.77,
2369
+ "grad_norm": 0.49148476123809814,
2370
+ "learning_rate": 2.6769129209986322e-05,
2371
+ "loss": 1.4618,
2372
+ "step": 333
2373
+ },
2374
+ {
2375
+ "epoch": 0.77,
2376
+ "grad_norm": 0.4833587408065796,
2377
+ "learning_rate": 2.6250753136210983e-05,
2378
+ "loss": 0.8573,
2379
+ "step": 334
2380
+ },
2381
+ {
2382
+ "epoch": 0.78,
2383
+ "grad_norm": 0.40724459290504456,
2384
+ "learning_rate": 2.5736686013646228e-05,
2385
+ "loss": 1.1546,
2386
+ "step": 335
2387
+ },
2388
+ {
2389
+ "epoch": 0.78,
2390
+ "grad_norm": 0.683252215385437,
2391
+ "learning_rate": 2.5226957877719436e-05,
2392
+ "loss": 1.1321,
2393
+ "step": 336
2394
+ },
2395
+ {
2396
+ "epoch": 0.78,
2397
+ "grad_norm": 0.6411616802215576,
2398
+ "learning_rate": 2.4721598510343858e-05,
2399
+ "loss": 1.3481,
2400
+ "step": 337
2401
+ },
2402
+ {
2403
+ "epoch": 0.78,
2404
+ "grad_norm": 0.4726378917694092,
2405
+ "learning_rate": 2.4220637438178317e-05,
2406
+ "loss": 1.0497,
2407
+ "step": 338
2408
+ },
2409
+ {
2410
+ "epoch": 0.79,
2411
+ "grad_norm": 0.6027311682701111,
2412
+ "learning_rate": 2.372410393090243e-05,
2413
+ "loss": 1.1339,
2414
+ "step": 339
2415
+ },
2416
+ {
2417
+ "epoch": 0.79,
2418
+ "grad_norm": 0.4566133916378021,
2419
+ "learning_rate": 2.3232026999506062e-05,
2420
+ "loss": 1.2384,
2421
+ "step": 340
2422
+ },
2423
+ {
2424
+ "epoch": 0.79,
2425
+ "grad_norm": 0.4425651431083679,
2426
+ "learning_rate": 2.2744435394594497e-05,
2427
+ "loss": 1.2542,
2428
+ "step": 341
2429
+ },
2430
+ {
2431
+ "epoch": 0.79,
2432
+ "grad_norm": 0.3810960352420807,
2433
+ "learning_rate": 2.22613576047087e-05,
2434
+ "loss": 0.9794,
2435
+ "step": 342
2436
+ },
2437
+ {
2438
+ "epoch": 0.8,
2439
+ "grad_norm": 0.441679447889328,
2440
+ "learning_rate": 2.1782821854660606e-05,
2441
+ "loss": 1.0967,
2442
+ "step": 343
2443
+ },
2444
+ {
2445
+ "epoch": 0.8,
2446
+ "grad_norm": 0.5362725257873535,
2447
+ "learning_rate": 2.130885610388428e-05,
2448
+ "loss": 1.3365,
2449
+ "step": 344
2450
+ },
2451
+ {
2452
+ "epoch": 0.8,
2453
+ "grad_norm": 0.5472970604896545,
2454
+ "learning_rate": 2.0839488044802036e-05,
2455
+ "loss": 1.7679,
2456
+ "step": 345
2457
+ },
2458
+ {
2459
+ "epoch": 0.8,
2460
+ "grad_norm": 0.4027257561683655,
2461
+ "learning_rate": 2.037474510120676e-05,
2462
+ "loss": 1.1387,
2463
+ "step": 346
2464
+ },
2465
+ {
2466
+ "epoch": 0.81,
2467
+ "grad_norm": 0.5073655843734741,
2468
+ "learning_rate": 1.9914654426659374e-05,
2469
+ "loss": 1.0684,
2470
+ "step": 347
2471
+ },
2472
+ {
2473
+ "epoch": 0.81,
2474
+ "grad_norm": 0.4431632161140442,
2475
+ "learning_rate": 1.945924290290242e-05,
2476
+ "loss": 1.1118,
2477
+ "step": 348
2478
+ },
2479
+ {
2480
+ "epoch": 0.81,
2481
+ "grad_norm": 0.7360476851463318,
2482
+ "learning_rate": 1.9008537138289527e-05,
2483
+ "loss": 1.5517,
2484
+ "step": 349
2485
+ },
2486
+ {
2487
+ "epoch": 0.81,
2488
+ "grad_norm": 0.43251582980155945,
2489
+ "learning_rate": 1.8562563466230576e-05,
2490
+ "loss": 1.1982,
2491
+ "step": 350
2492
+ },
2493
+ {
2494
+ "epoch": 0.81,
2495
+ "grad_norm": 0.4459226429462433,
2496
+ "learning_rate": 1.8121347943653332e-05,
2497
+ "loss": 1.3095,
2498
+ "step": 351
2499
+ },
2500
+ {
2501
+ "epoch": 0.82,
2502
+ "grad_norm": 0.34792765974998474,
2503
+ "learning_rate": 1.7684916349480794e-05,
2504
+ "loss": 0.9415,
2505
+ "step": 352
2506
+ },
2507
+ {
2508
+ "epoch": 0.82,
2509
+ "grad_norm": 0.5468625426292419,
2510
+ "learning_rate": 1.7253294183125223e-05,
2511
+ "loss": 1.1621,
2512
+ "step": 353
2513
+ },
2514
+ {
2515
+ "epoch": 0.82,
2516
+ "grad_norm": 0.5048443675041199,
2517
+ "learning_rate": 1.6826506662998097e-05,
2518
+ "loss": 1.3502,
2519
+ "step": 354
2520
+ },
2521
+ {
2522
+ "epoch": 0.82,
2523
+ "grad_norm": 0.3652397692203522,
2524
+ "learning_rate": 1.64045787250368e-05,
2525
+ "loss": 1.1079,
2526
+ "step": 355
2527
+ },
2528
+ {
2529
+ "epoch": 0.83,
2530
+ "grad_norm": 0.45304709672927856,
2531
+ "learning_rate": 1.5987535021247667e-05,
2532
+ "loss": 1.0239,
2533
+ "step": 356
2534
+ },
2535
+ {
2536
+ "epoch": 0.83,
2537
+ "grad_norm": 0.38547682762145996,
2538
+ "learning_rate": 1.5575399918265542e-05,
2539
+ "loss": 1.1345,
2540
+ "step": 357
2541
+ },
2542
+ {
2543
+ "epoch": 0.83,
2544
+ "grad_norm": 0.47186747193336487,
2545
+ "learning_rate": 1.5168197495930315e-05,
2546
+ "loss": 1.4572,
2547
+ "step": 358
2548
+ },
2549
+ {
2550
+ "epoch": 0.83,
2551
+ "grad_norm": 0.4306280314922333,
2552
+ "learning_rate": 1.476595154587973e-05,
2553
+ "loss": 0.9894,
2554
+ "step": 359
2555
+ },
2556
+ {
2557
+ "epoch": 0.84,
2558
+ "grad_norm": 0.49191519618034363,
2559
+ "learning_rate": 1.436868557015959e-05,
2560
+ "loss": 1.3018,
2561
+ "step": 360
2562
+ },
2563
+ {
2564
+ "epoch": 0.84,
2565
+ "grad_norm": 0.548028826713562,
2566
+ "learning_rate": 1.3976422779850384e-05,
2567
+ "loss": 1.3219,
2568
+ "step": 361
2569
+ },
2570
+ {
2571
+ "epoch": 0.84,
2572
+ "grad_norm": 0.39407745003700256,
2573
+ "learning_rate": 1.3589186093711226e-05,
2574
+ "loss": 1.002,
2575
+ "step": 362
2576
+ },
2577
+ {
2578
+ "epoch": 0.84,
2579
+ "grad_norm": 0.47022268176078796,
2580
+ "learning_rate": 1.3206998136840831e-05,
2581
+ "loss": 1.2112,
2582
+ "step": 363
2583
+ },
2584
+ {
2585
+ "epoch": 0.84,
2586
+ "grad_norm": 0.7266699075698853,
2587
+ "learning_rate": 1.2829881239355468e-05,
2588
+ "loss": 1.3021,
2589
+ "step": 364
2590
+ },
2591
+ {
2592
+ "epoch": 0.85,
2593
+ "grad_norm": 0.6192635297775269,
2594
+ "learning_rate": 1.2457857435084408e-05,
2595
+ "loss": 2.7741,
2596
+ "step": 365
2597
+ },
2598
+ {
2599
+ "epoch": 0.85,
2600
+ "grad_norm": 0.39360511302948,
2601
+ "learning_rate": 1.2090948460282414e-05,
2602
+ "loss": 1.1131,
2603
+ "step": 366
2604
+ },
2605
+ {
2606
+ "epoch": 0.85,
2607
+ "grad_norm": 0.6498689651489258,
2608
+ "learning_rate": 1.1729175752359922e-05,
2609
+ "loss": 1.3666,
2610
+ "step": 367
2611
+ },
2612
+ {
2613
+ "epoch": 0.85,
2614
+ "grad_norm": 0.40655967593193054,
2615
+ "learning_rate": 1.1372560448630376e-05,
2616
+ "loss": 1.1841,
2617
+ "step": 368
2618
+ },
2619
+ {
2620
+ "epoch": 0.86,
2621
+ "grad_norm": 0.3894653022289276,
2622
+ "learning_rate": 1.102112338507526e-05,
2623
+ "loss": 1.1494,
2624
+ "step": 369
2625
+ },
2626
+ {
2627
+ "epoch": 0.86,
2628
+ "grad_norm": 0.5716227889060974,
2629
+ "learning_rate": 1.067488509512683e-05,
2630
+ "loss": 0.9971,
2631
+ "step": 370
2632
+ },
2633
+ {
2634
+ "epoch": 0.86,
2635
+ "grad_norm": 0.4679757356643677,
2636
+ "learning_rate": 1.0333865808468202e-05,
2637
+ "loss": 1.0505,
2638
+ "step": 371
2639
+ },
2640
+ {
2641
+ "epoch": 0.86,
2642
+ "grad_norm": 0.5550671219825745,
2643
+ "learning_rate": 9.998085449851635e-06,
2644
+ "loss": 1.0006,
2645
+ "step": 372
2646
+ },
2647
+ {
2648
+ "epoch": 0.87,
2649
+ "grad_norm": 0.4179516136646271,
2650
+ "learning_rate": 9.667563637934129e-06,
2651
+ "loss": 0.8996,
2652
+ "step": 373
2653
+ },
2654
+ {
2655
+ "epoch": 0.87,
2656
+ "grad_norm": 0.4545114040374756,
2657
+ "learning_rate": 9.342319684131395e-06,
2658
+ "loss": 1.1175,
2659
+ "step": 374
2660
+ },
2661
+ {
2662
+ "epoch": 0.87,
2663
+ "grad_norm": 0.5175038576126099,
2664
+ "learning_rate": 9.02237259148938e-06,
2665
+ "loss": 1.1729,
2666
+ "step": 375
2667
+ },
2668
+ {
2669
+ "epoch": 0.87,
2670
+ "grad_norm": 0.45041874051094055,
2671
+ "learning_rate": 8.70774105357407e-06,
2672
+ "loss": 1.2005,
2673
+ "step": 376
2674
+ },
2675
+ {
2676
+ "epoch": 0.87,
2677
+ "grad_norm": 0.47179684042930603,
2678
+ "learning_rate": 8.398443453379267e-06,
2679
+ "loss": 1.1232,
2680
+ "step": 377
2681
+ },
2682
+ {
2683
+ "epoch": 0.88,
2684
+ "grad_norm": 0.4453977942466736,
2685
+ "learning_rate": 8.094497862252471e-06,
2686
+ "loss": 1.3498,
2687
+ "step": 378
2688
+ },
2689
+ {
2690
+ "epoch": 0.88,
2691
+ "grad_norm": 0.545090913772583,
2692
+ "learning_rate": 7.795922038839032e-06,
2693
+ "loss": 1.3135,
2694
+ "step": 379
2695
+ },
2696
+ {
2697
+ "epoch": 0.88,
2698
+ "grad_norm": 0.41150522232055664,
2699
+ "learning_rate": 7.502733428044683e-06,
2700
+ "loss": 1.0358,
2701
+ "step": 380
2702
+ },
2703
+ {
2704
+ "epoch": 0.88,
2705
+ "grad_norm": 0.4423467218875885,
2706
+ "learning_rate": 7.214949160016115e-06,
2707
+ "loss": 1.1287,
2708
+ "step": 381
2709
+ },
2710
+ {
2711
+ "epoch": 0.89,
2712
+ "grad_norm": 0.4803576171398163,
2713
+ "learning_rate": 6.932586049140255e-06,
2714
+ "loss": 1.1248,
2715
+ "step": 382
2716
+ },
2717
+ {
2718
+ "epoch": 0.89,
2719
+ "grad_norm": 0.50335294008255,
2720
+ "learning_rate": 6.655660593061719e-06,
2721
+ "loss": 1.2563,
2722
+ "step": 383
2723
+ },
2724
+ {
2725
+ "epoch": 0.89,
2726
+ "grad_norm": 0.4255715310573578,
2727
+ "learning_rate": 6.384188971719052e-06,
2728
+ "loss": 1.1771,
2729
+ "step": 384
2730
+ },
2731
+ {
2732
+ "epoch": 0.89,
2733
+ "grad_norm": 0.5371767282485962,
2734
+ "learning_rate": 6.11818704639926e-06,
2735
+ "loss": 1.4661,
2736
+ "step": 385
2737
+ },
2738
+ {
2739
+ "epoch": 0.9,
2740
+ "grad_norm": 0.7055171132087708,
2741
+ "learning_rate": 5.857670358811096e-06,
2742
+ "loss": 1.2068,
2743
+ "step": 386
2744
+ },
2745
+ {
2746
+ "epoch": 0.9,
2747
+ "grad_norm": 0.6857877969741821,
2748
+ "learning_rate": 5.6026541301771095e-06,
2749
+ "loss": 1.025,
2750
+ "step": 387
2751
+ },
2752
+ {
2753
+ "epoch": 0.9,
2754
+ "grad_norm": 0.3706812858581543,
2755
+ "learning_rate": 5.353153260344179e-06,
2756
+ "loss": 0.4216,
2757
+ "step": 388
2758
+ },
2759
+ {
2760
+ "epoch": 0.9,
2761
+ "grad_norm": 0.3943740725517273,
2762
+ "learning_rate": 5.109182326913054e-06,
2763
+ "loss": 1.1772,
2764
+ "step": 389
2765
+ },
2766
+ {
2767
+ "epoch": 0.9,
2768
+ "grad_norm": 0.44216853380203247,
2769
+ "learning_rate": 4.870755584386544e-06,
2770
+ "loss": 1.1343,
2771
+ "step": 390
2772
+ },
2773
+ {
2774
+ "epoch": 0.91,
2775
+ "grad_norm": 0.5114327669143677,
2776
+ "learning_rate": 4.63788696333678e-06,
2777
+ "loss": 1.3096,
2778
+ "step": 391
2779
+ },
2780
+ {
2781
+ "epoch": 0.91,
2782
+ "grad_norm": 0.4400082230567932,
2783
+ "learning_rate": 4.410590069591192e-06,
2784
+ "loss": 1.0596,
2785
+ "step": 392
2786
+ },
2787
+ {
2788
+ "epoch": 0.91,
2789
+ "grad_norm": 0.5141562223434448,
2790
+ "learning_rate": 4.188878183437594e-06,
2791
+ "loss": 0.976,
2792
+ "step": 393
2793
+ },
2794
+ {
2795
+ "epoch": 0.91,
2796
+ "grad_norm": 0.47619250416755676,
2797
+ "learning_rate": 3.972764258848305e-06,
2798
+ "loss": 0.8015,
2799
+ "step": 394
2800
+ },
2801
+ {
2802
+ "epoch": 0.92,
2803
+ "grad_norm": 0.44292959570884705,
2804
+ "learning_rate": 3.7622609227231818e-06,
2805
+ "loss": 1.2856,
2806
+ "step": 395
2807
+ },
2808
+ {
2809
+ "epoch": 0.92,
2810
+ "grad_norm": 0.5242791175842285,
2811
+ "learning_rate": 3.5573804741519833e-06,
2812
+ "loss": 1.0283,
2813
+ "step": 396
2814
+ },
2815
+ {
2816
+ "epoch": 0.92,
2817
+ "grad_norm": 0.41017472743988037,
2818
+ "learning_rate": 3.3581348836956738e-06,
2819
+ "loss": 0.8387,
2820
+ "step": 397
2821
+ },
2822
+ {
2823
+ "epoch": 0.92,
2824
+ "grad_norm": 0.44496363401412964,
2825
+ "learning_rate": 3.1645357926870955e-06,
2826
+ "loss": 1.0878,
2827
+ "step": 398
2828
+ },
2829
+ {
2830
+ "epoch": 0.93,
2831
+ "grad_norm": 0.5821820497512817,
2832
+ "learning_rate": 2.9765945125507235e-06,
2833
+ "loss": 1.3477,
2834
+ "step": 399
2835
+ },
2836
+ {
2837
+ "epoch": 0.93,
2838
+ "grad_norm": 0.5164168477058411,
2839
+ "learning_rate": 2.7943220241418377e-06,
2840
+ "loss": 1.6242,
2841
+ "step": 400
2842
+ },
2843
+ {
2844
+ "epoch": 0.93,
2845
+ "grad_norm": 0.5002605319023132,
2846
+ "learning_rate": 2.6177289771049274e-06,
2847
+ "loss": 1.2107,
2848
+ "step": 401
2849
+ },
2850
+ {
2851
+ "epoch": 0.93,
2852
+ "grad_norm": 0.44040682911872864,
2853
+ "learning_rate": 2.4468256892514417e-06,
2854
+ "loss": 0.8678,
2855
+ "step": 402
2856
+ },
2857
+ {
2858
+ "epoch": 0.94,
2859
+ "grad_norm": 0.35371842980384827,
2860
+ "learning_rate": 2.281622145956952e-06,
2861
+ "loss": 0.7567,
2862
+ "step": 403
2863
+ },
2864
+ {
2865
+ "epoch": 0.94,
2866
+ "grad_norm": 0.4473396837711334,
2867
+ "learning_rate": 2.122127999577783e-06,
2868
+ "loss": 1.0541,
2869
+ "step": 404
2870
+ },
2871
+ {
2872
+ "epoch": 0.94,
2873
+ "grad_norm": 0.6327199339866638,
2874
+ "learning_rate": 1.9683525688869773e-06,
2875
+ "loss": 1.2364,
2876
+ "step": 405
2877
+ },
2878
+ {
2879
+ "epoch": 0.94,
2880
+ "grad_norm": 0.4338397681713104,
2881
+ "learning_rate": 1.8203048385299181e-06,
2882
+ "loss": 1.1542,
2883
+ "step": 406
2884
+ },
2885
+ {
2886
+ "epoch": 0.94,
2887
+ "grad_norm": 0.5743327140808105,
2888
+ "learning_rate": 1.6779934584992718e-06,
2889
+ "loss": 1.2498,
2890
+ "step": 407
2891
+ },
2892
+ {
2893
+ "epoch": 0.95,
2894
+ "grad_norm": 0.40991920232772827,
2895
+ "learning_rate": 1.5414267436297037e-06,
2896
+ "loss": 1.1019,
2897
+ "step": 408
2898
+ },
2899
+ {
2900
+ "epoch": 0.95,
2901
+ "grad_norm": 0.40616923570632935,
2902
+ "learning_rate": 1.4106126731119996e-06,
2903
+ "loss": 1.2456,
2904
+ "step": 409
2905
+ },
2906
+ {
2907
+ "epoch": 0.95,
2908
+ "grad_norm": 0.4829423725605011,
2909
+ "learning_rate": 1.2855588900269056e-06,
2910
+ "loss": 1.1966,
2911
+ "step": 410
2912
+ },
2913
+ {
2914
+ "epoch": 0.95,
2915
+ "grad_norm": 0.4857659339904785,
2916
+ "learning_rate": 1.1662727008984964e-06,
2917
+ "loss": 1.4712,
2918
+ "step": 411
2919
+ },
2920
+ {
2921
+ "epoch": 0.96,
2922
+ "grad_norm": 0.4441464841365814,
2923
+ "learning_rate": 1.0527610752673944e-06,
2924
+ "loss": 1.1358,
2925
+ "step": 412
2926
+ },
2927
+ {
2928
+ "epoch": 0.96,
2929
+ "grad_norm": 0.4043671488761902,
2930
+ "learning_rate": 9.450306452834179e-07,
2931
+ "loss": 1.1369,
2932
+ "step": 413
2933
+ },
2934
+ {
2935
+ "epoch": 0.96,
2936
+ "grad_norm": 0.5048549771308899,
2937
+ "learning_rate": 8.430877053182129e-07,
2938
+ "loss": 1.1279,
2939
+ "step": 414
2940
+ },
2941
+ {
2942
+ "epoch": 0.96,
2943
+ "grad_norm": 0.43177226185798645,
2944
+ "learning_rate": 7.469382115974032e-07,
2945
+ "loss": 1.269,
2946
+ "step": 415
2947
+ },
2948
+ {
2949
+ "epoch": 0.97,
2950
+ "grad_norm": 0.45105987787246704,
2951
+ "learning_rate": 6.565877818526245e-07,
2952
+ "loss": 1.6302,
2953
+ "step": 416
2954
+ },
2955
+ {
2956
+ "epoch": 0.97,
2957
+ "grad_norm": 0.4008980989456177,
2958
+ "learning_rate": 5.72041694993286e-07,
2959
+ "loss": 0.7278,
2960
+ "step": 417
2961
+ },
2962
+ {
2963
+ "epoch": 0.97,
2964
+ "grad_norm": 0.570662796497345,
2965
+ "learning_rate": 4.933048907981741e-07,
2966
+ "loss": 1.2836,
2967
+ "step": 418
2968
+ },
2969
+ {
2970
+ "epoch": 0.97,
2971
+ "grad_norm": 0.39943748712539673,
2972
+ "learning_rate": 4.203819696267486e-07,
2973
+ "loss": 1.1154,
2974
+ "step": 419
2975
+ },
2976
+ {
2977
+ "epoch": 0.97,
2978
+ "grad_norm": 0.6464762687683105,
2979
+ "learning_rate": 3.532771921504696e-07,
2980
+ "loss": 0.9646,
2981
+ "step": 420
2982
+ },
2983
+ {
2984
+ "epoch": 0.98,
2985
+ "grad_norm": 0.46713489294052124,
2986
+ "learning_rate": 2.919944791037632e-07,
2987
+ "loss": 1.1209,
2988
+ "step": 421
2989
+ },
2990
+ {
2991
+ "epoch": 0.98,
2992
+ "grad_norm": 0.45911917090415955,
2993
+ "learning_rate": 2.3653741105499338e-07,
2994
+ "loss": 1.2976,
2995
+ "step": 422
2996
+ },
2997
+ {
2998
+ "epoch": 0.98,
2999
+ "grad_norm": 0.4376751482486725,
3000
+ "learning_rate": 1.8690922819727398e-07,
3001
+ "loss": 1.2812,
3002
+ "step": 423
3003
+ },
3004
+ {
3005
+ "epoch": 0.98,
3006
+ "grad_norm": 0.461121529340744,
3007
+ "learning_rate": 1.4311283015910893e-07,
3008
+ "loss": 1.4073,
3009
+ "step": 424
3010
+ },
3011
+ {
3012
+ "epoch": 0.99,
3013
+ "grad_norm": 0.39589831233024597,
3014
+ "learning_rate": 1.0515077583498344e-07,
3015
+ "loss": 1.0048,
3016
+ "step": 425
3017
+ },
3018
+ {
3019
+ "epoch": 0.99,
3020
+ "grad_norm": 0.39896348118782043,
3021
+ "learning_rate": 7.302528323589464e-08,
3022
+ "loss": 0.8817,
3023
+ "step": 426
3024
+ },
3025
+ {
3026
+ "epoch": 0.99,
3027
+ "grad_norm": 0.43354111909866333,
3028
+ "learning_rate": 4.6738229359732935e-08,
3029
+ "loss": 0.9946,
3030
+ "step": 427
3031
+ },
3032
+ {
3033
+ "epoch": 0.99,
3034
+ "grad_norm": 0.3909148871898651,
3035
+ "learning_rate": 2.6291150081603212e-08,
3036
+ "loss": 1.1572,
3037
+ "step": 428
3038
+ },
3039
+ {
3040
+ "epoch": 1.0,
3041
+ "grad_norm": 0.5046055912971497,
3042
+ "learning_rate": 1.168524006410765e-08,
3043
+ "loss": 1.3007,
3044
+ "step": 429
3045
+ },
3046
+ {
3047
+ "epoch": 1.0,
3048
+ "grad_norm": 0.6245082020759583,
3049
+ "learning_rate": 2.921352687534906e-09,
3050
+ "loss": 0.9887,
3051
+ "step": 430
3052
+ },
3053
+ {
3054
+ "epoch": 1.0,
3055
+ "grad_norm": 0.45793771743774414,
3056
+ "learning_rate": 0.0,
3057
+ "loss": 1.1817,
3058
+ "step": 431
3059
+ }
3060
+ ],
3061
+ "logging_steps": 1,
3062
+ "max_steps": 431,
3063
+ "num_input_tokens_seen": 0,
3064
+ "num_train_epochs": 1,
3065
+ "save_steps": 500,
3066
+ "total_flos": 7928752835297280.0,
3067
+ "train_batch_size": 1,
3068
+ "trial_name": null,
3069
+ "trial_params": null
3070
+ }
checkpoint-431/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d319bb99efc5ca0c69c268d4e6294eb7365187779ec9c45fc4ebc1cefa3e0c93
3
+ size 5624
config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openlm-research/open_llama_3b_v2",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 3200,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8640,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 26,
18
+ "num_key_value_heads": 32,
19
+ "pad_token_id": 0,
20
+ "pretraining_tp": 1,
21
+ "quantization_config": {
22
+ "_load_in_4bit": true,
23
+ "_load_in_8bit": false,
24
+ "bnb_4bit_compute_dtype": "float16",
25
+ "bnb_4bit_quant_type": "nf4",
26
+ "bnb_4bit_use_double_quant": true,
27
+ "llm_int8_enable_fp32_cpu_offload": false,
28
+ "llm_int8_has_fp16_weight": false,
29
+ "llm_int8_skip_modules": null,
30
+ "llm_int8_threshold": 6.0,
31
+ "load_in_4bit": true,
32
+ "load_in_8bit": false,
33
+ "quant_method": "bitsandbytes"
34
+ },
35
+ "rms_norm_eps": 1e-06,
36
+ "rope_scaling": null,
37
+ "rope_theta": 10000.0,
38
+ "tie_word_embeddings": false,
39
+ "torch_dtype": "float16",
40
+ "transformers_version": "4.38.2",
41
+ "use_cache": false,
42
+ "vocab_size": 32000
43
+ }
runs/Mar08_18-22-13_80d2e4e4a3f6/events.out.tfevents.1709922133.80d2e4e4a3f6.198.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06e290263708d22d425c8fa94b2c4267b14a9f631bc33709151664e68e662b46
3
+ size 11039
runs/Mar08_18-28-26_80d2e4e4a3f6/events.out.tfevents.1709922507.80d2e4e4a3f6.226.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b882cfe87aaf7fa1b09aa5b271cf5719299a20b0048fbf6bdbbf1cbfeb7ddab2
3
+ size 97666
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b289e85fa20fd375d8b33dc12f77616f18abc6359804471d1fafcb425fecb8
3
+ size 511574
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 2048,
36
+ "pad_token": "</s>",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "use_fast": true
43
+ }