Apertus-8B-Instruct-2509-exl3 / quantization_config.json
turboderp's picture
Upload folder using huggingface_hub
c44de1d verified
raw
history blame
230 kB
{
"quant_method": "exl3",
"version": "0.0.6",
"bits": 4.0,
"head_bits": 6,
"calibration": {
"rows": 100,
"cols": 2048
},
"out_scales": "auto",
"tensor_storage": {
"model.embed_tokens": {
"stored_tensors": {
"model.embed_tokens.weight": {
"shape": [
131072,
4096
],
"n_bytes": 1073741824,
"dtype": "torch.float16"
}
}
},
"model.layers.0.attention_layernorm": {
"stored_tensors": {
"model.layers.0.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.0.self_attn.q_proj": {
"stored_tensors": {
"model.layers.0.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.0.self_attn.k_proj": {
"stored_tensors": {
"model.layers.0.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.0.self_attn.v_proj": {
"stored_tensors": {
"model.layers.0.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.0.self_attn.o_proj": {
"stored_tensors": {
"model.layers.0.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.0.self_attn.q_norm": {
"stored_tensors": {
"model.layers.0.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.0.self_attn.k_norm": {
"stored_tensors": {
"model.layers.0.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.0.feedforward_layernorm": {
"stored_tensors": {
"model.layers.0.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.0.mlp.up_proj": {
"stored_tensors": {
"model.layers.0.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.0.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.0.mlp.down_proj": {
"stored_tensors": {
"model.layers.0.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.0.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.0.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.attention_layernorm": {
"stored_tensors": {
"model.layers.1.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.1.self_attn.q_proj": {
"stored_tensors": {
"model.layers.1.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.self_attn.k_proj": {
"stored_tensors": {
"model.layers.1.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.self_attn.v_proj": {
"stored_tensors": {
"model.layers.1.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.self_attn.o_proj": {
"stored_tensors": {
"model.layers.1.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.self_attn.q_norm": {
"stored_tensors": {
"model.layers.1.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.1.self_attn.k_norm": {
"stored_tensors": {
"model.layers.1.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.1.feedforward_layernorm": {
"stored_tensors": {
"model.layers.1.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.1.mlp.up_proj": {
"stored_tensors": {
"model.layers.1.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.1.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.1.mlp.down_proj": {
"stored_tensors": {
"model.layers.1.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.1.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.1.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.attention_layernorm": {
"stored_tensors": {
"model.layers.2.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.2.self_attn.q_proj": {
"stored_tensors": {
"model.layers.2.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.self_attn.k_proj": {
"stored_tensors": {
"model.layers.2.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.self_attn.v_proj": {
"stored_tensors": {
"model.layers.2.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.self_attn.o_proj": {
"stored_tensors": {
"model.layers.2.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.self_attn.q_norm": {
"stored_tensors": {
"model.layers.2.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.2.self_attn.k_norm": {
"stored_tensors": {
"model.layers.2.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.2.feedforward_layernorm": {
"stored_tensors": {
"model.layers.2.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.2.mlp.up_proj": {
"stored_tensors": {
"model.layers.2.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.2.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.2.mlp.down_proj": {
"stored_tensors": {
"model.layers.2.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.2.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.2.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.attention_layernorm": {
"stored_tensors": {
"model.layers.3.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.3.self_attn.q_proj": {
"stored_tensors": {
"model.layers.3.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.self_attn.k_proj": {
"stored_tensors": {
"model.layers.3.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.self_attn.v_proj": {
"stored_tensors": {
"model.layers.3.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.self_attn.o_proj": {
"stored_tensors": {
"model.layers.3.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.self_attn.q_norm": {
"stored_tensors": {
"model.layers.3.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.3.self_attn.k_norm": {
"stored_tensors": {
"model.layers.3.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.3.feedforward_layernorm": {
"stored_tensors": {
"model.layers.3.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.3.mlp.up_proj": {
"stored_tensors": {
"model.layers.3.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.3.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.3.mlp.down_proj": {
"stored_tensors": {
"model.layers.3.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.3.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.3.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.attention_layernorm": {
"stored_tensors": {
"model.layers.4.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.4.self_attn.q_proj": {
"stored_tensors": {
"model.layers.4.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.self_attn.k_proj": {
"stored_tensors": {
"model.layers.4.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.self_attn.v_proj": {
"stored_tensors": {
"model.layers.4.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.self_attn.o_proj": {
"stored_tensors": {
"model.layers.4.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.self_attn.q_norm": {
"stored_tensors": {
"model.layers.4.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.4.self_attn.k_norm": {
"stored_tensors": {
"model.layers.4.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.4.feedforward_layernorm": {
"stored_tensors": {
"model.layers.4.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.4.mlp.up_proj": {
"stored_tensors": {
"model.layers.4.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.4.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.4.mlp.down_proj": {
"stored_tensors": {
"model.layers.4.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.4.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.4.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.attention_layernorm": {
"stored_tensors": {
"model.layers.5.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.5.self_attn.q_proj": {
"stored_tensors": {
"model.layers.5.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.self_attn.k_proj": {
"stored_tensors": {
"model.layers.5.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.self_attn.v_proj": {
"stored_tensors": {
"model.layers.5.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.self_attn.o_proj": {
"stored_tensors": {
"model.layers.5.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.self_attn.q_norm": {
"stored_tensors": {
"model.layers.5.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.5.self_attn.k_norm": {
"stored_tensors": {
"model.layers.5.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.5.feedforward_layernorm": {
"stored_tensors": {
"model.layers.5.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.5.mlp.up_proj": {
"stored_tensors": {
"model.layers.5.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.5.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.5.mlp.down_proj": {
"stored_tensors": {
"model.layers.5.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.5.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.5.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.attention_layernorm": {
"stored_tensors": {
"model.layers.6.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.6.self_attn.q_proj": {
"stored_tensors": {
"model.layers.6.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.self_attn.k_proj": {
"stored_tensors": {
"model.layers.6.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.self_attn.v_proj": {
"stored_tensors": {
"model.layers.6.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.self_attn.o_proj": {
"stored_tensors": {
"model.layers.6.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.self_attn.q_norm": {
"stored_tensors": {
"model.layers.6.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.6.self_attn.k_norm": {
"stored_tensors": {
"model.layers.6.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.6.feedforward_layernorm": {
"stored_tensors": {
"model.layers.6.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.6.mlp.up_proj": {
"stored_tensors": {
"model.layers.6.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.6.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.6.mlp.down_proj": {
"stored_tensors": {
"model.layers.6.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.6.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.6.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.attention_layernorm": {
"stored_tensors": {
"model.layers.7.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.7.self_attn.q_proj": {
"stored_tensors": {
"model.layers.7.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.self_attn.k_proj": {
"stored_tensors": {
"model.layers.7.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.self_attn.v_proj": {
"stored_tensors": {
"model.layers.7.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.self_attn.o_proj": {
"stored_tensors": {
"model.layers.7.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.self_attn.q_norm": {
"stored_tensors": {
"model.layers.7.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.7.self_attn.k_norm": {
"stored_tensors": {
"model.layers.7.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.7.feedforward_layernorm": {
"stored_tensors": {
"model.layers.7.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.7.mlp.up_proj": {
"stored_tensors": {
"model.layers.7.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.7.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.7.mlp.down_proj": {
"stored_tensors": {
"model.layers.7.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.7.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.7.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.attention_layernorm": {
"stored_tensors": {
"model.layers.8.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.8.self_attn.q_proj": {
"stored_tensors": {
"model.layers.8.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.self_attn.k_proj": {
"stored_tensors": {
"model.layers.8.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.self_attn.v_proj": {
"stored_tensors": {
"model.layers.8.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.self_attn.o_proj": {
"stored_tensors": {
"model.layers.8.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.self_attn.q_norm": {
"stored_tensors": {
"model.layers.8.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.8.self_attn.k_norm": {
"stored_tensors": {
"model.layers.8.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.8.feedforward_layernorm": {
"stored_tensors": {
"model.layers.8.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.8.mlp.up_proj": {
"stored_tensors": {
"model.layers.8.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.8.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.8.mlp.down_proj": {
"stored_tensors": {
"model.layers.8.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.8.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.8.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.attention_layernorm": {
"stored_tensors": {
"model.layers.9.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.9.self_attn.q_proj": {
"stored_tensors": {
"model.layers.9.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.self_attn.k_proj": {
"stored_tensors": {
"model.layers.9.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.self_attn.v_proj": {
"stored_tensors": {
"model.layers.9.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.self_attn.o_proj": {
"stored_tensors": {
"model.layers.9.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.self_attn.q_norm": {
"stored_tensors": {
"model.layers.9.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.9.self_attn.k_norm": {
"stored_tensors": {
"model.layers.9.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.9.feedforward_layernorm": {
"stored_tensors": {
"model.layers.9.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.9.mlp.up_proj": {
"stored_tensors": {
"model.layers.9.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.9.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.9.mlp.down_proj": {
"stored_tensors": {
"model.layers.9.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.9.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.9.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.attention_layernorm": {
"stored_tensors": {
"model.layers.10.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.10.self_attn.q_proj": {
"stored_tensors": {
"model.layers.10.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.self_attn.k_proj": {
"stored_tensors": {
"model.layers.10.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.self_attn.v_proj": {
"stored_tensors": {
"model.layers.10.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.self_attn.o_proj": {
"stored_tensors": {
"model.layers.10.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.self_attn.q_norm": {
"stored_tensors": {
"model.layers.10.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.10.self_attn.k_norm": {
"stored_tensors": {
"model.layers.10.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.10.feedforward_layernorm": {
"stored_tensors": {
"model.layers.10.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.10.mlp.up_proj": {
"stored_tensors": {
"model.layers.10.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.10.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.10.mlp.down_proj": {
"stored_tensors": {
"model.layers.10.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.10.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.10.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.attention_layernorm": {
"stored_tensors": {
"model.layers.11.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.11.self_attn.q_proj": {
"stored_tensors": {
"model.layers.11.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.self_attn.k_proj": {
"stored_tensors": {
"model.layers.11.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.self_attn.v_proj": {
"stored_tensors": {
"model.layers.11.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.self_attn.o_proj": {
"stored_tensors": {
"model.layers.11.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.self_attn.q_norm": {
"stored_tensors": {
"model.layers.11.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.11.self_attn.k_norm": {
"stored_tensors": {
"model.layers.11.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.11.feedforward_layernorm": {
"stored_tensors": {
"model.layers.11.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.11.mlp.up_proj": {
"stored_tensors": {
"model.layers.11.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.11.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.11.mlp.down_proj": {
"stored_tensors": {
"model.layers.11.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.11.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.11.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.attention_layernorm": {
"stored_tensors": {
"model.layers.12.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.12.self_attn.q_proj": {
"stored_tensors": {
"model.layers.12.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.self_attn.k_proj": {
"stored_tensors": {
"model.layers.12.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.self_attn.v_proj": {
"stored_tensors": {
"model.layers.12.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.self_attn.o_proj": {
"stored_tensors": {
"model.layers.12.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.self_attn.q_norm": {
"stored_tensors": {
"model.layers.12.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.12.self_attn.k_norm": {
"stored_tensors": {
"model.layers.12.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.12.feedforward_layernorm": {
"stored_tensors": {
"model.layers.12.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.12.mlp.up_proj": {
"stored_tensors": {
"model.layers.12.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.12.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.12.mlp.down_proj": {
"stored_tensors": {
"model.layers.12.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.12.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.12.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.attention_layernorm": {
"stored_tensors": {
"model.layers.13.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.13.self_attn.q_proj": {
"stored_tensors": {
"model.layers.13.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.self_attn.k_proj": {
"stored_tensors": {
"model.layers.13.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.self_attn.v_proj": {
"stored_tensors": {
"model.layers.13.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.self_attn.o_proj": {
"stored_tensors": {
"model.layers.13.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.self_attn.q_norm": {
"stored_tensors": {
"model.layers.13.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.13.self_attn.k_norm": {
"stored_tensors": {
"model.layers.13.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.13.feedforward_layernorm": {
"stored_tensors": {
"model.layers.13.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.13.mlp.up_proj": {
"stored_tensors": {
"model.layers.13.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.13.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.13.mlp.down_proj": {
"stored_tensors": {
"model.layers.13.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.13.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.13.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.attention_layernorm": {
"stored_tensors": {
"model.layers.14.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.14.self_attn.q_proj": {
"stored_tensors": {
"model.layers.14.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.self_attn.k_proj": {
"stored_tensors": {
"model.layers.14.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.self_attn.v_proj": {
"stored_tensors": {
"model.layers.14.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.self_attn.o_proj": {
"stored_tensors": {
"model.layers.14.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.self_attn.q_norm": {
"stored_tensors": {
"model.layers.14.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.14.self_attn.k_norm": {
"stored_tensors": {
"model.layers.14.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.14.feedforward_layernorm": {
"stored_tensors": {
"model.layers.14.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.14.mlp.up_proj": {
"stored_tensors": {
"model.layers.14.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.14.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.14.mlp.down_proj": {
"stored_tensors": {
"model.layers.14.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.14.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.14.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.attention_layernorm": {
"stored_tensors": {
"model.layers.15.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.15.self_attn.q_proj": {
"stored_tensors": {
"model.layers.15.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.self_attn.k_proj": {
"stored_tensors": {
"model.layers.15.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.self_attn.v_proj": {
"stored_tensors": {
"model.layers.15.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.self_attn.o_proj": {
"stored_tensors": {
"model.layers.15.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.self_attn.q_norm": {
"stored_tensors": {
"model.layers.15.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.15.self_attn.k_norm": {
"stored_tensors": {
"model.layers.15.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.15.feedforward_layernorm": {
"stored_tensors": {
"model.layers.15.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.15.mlp.up_proj": {
"stored_tensors": {
"model.layers.15.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.15.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.15.mlp.down_proj": {
"stored_tensors": {
"model.layers.15.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.15.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.15.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.attention_layernorm": {
"stored_tensors": {
"model.layers.16.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.16.self_attn.q_proj": {
"stored_tensors": {
"model.layers.16.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.self_attn.k_proj": {
"stored_tensors": {
"model.layers.16.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.self_attn.v_proj": {
"stored_tensors": {
"model.layers.16.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.self_attn.o_proj": {
"stored_tensors": {
"model.layers.16.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.self_attn.q_norm": {
"stored_tensors": {
"model.layers.16.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.16.self_attn.k_norm": {
"stored_tensors": {
"model.layers.16.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.16.feedforward_layernorm": {
"stored_tensors": {
"model.layers.16.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.16.mlp.up_proj": {
"stored_tensors": {
"model.layers.16.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.16.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.16.mlp.down_proj": {
"stored_tensors": {
"model.layers.16.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.16.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.16.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.attention_layernorm": {
"stored_tensors": {
"model.layers.17.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.17.self_attn.q_proj": {
"stored_tensors": {
"model.layers.17.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.self_attn.k_proj": {
"stored_tensors": {
"model.layers.17.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.self_attn.v_proj": {
"stored_tensors": {
"model.layers.17.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.self_attn.o_proj": {
"stored_tensors": {
"model.layers.17.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.self_attn.q_norm": {
"stored_tensors": {
"model.layers.17.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.17.self_attn.k_norm": {
"stored_tensors": {
"model.layers.17.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.17.feedforward_layernorm": {
"stored_tensors": {
"model.layers.17.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.17.mlp.up_proj": {
"stored_tensors": {
"model.layers.17.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.17.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.17.mlp.down_proj": {
"stored_tensors": {
"model.layers.17.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.17.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.17.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.attention_layernorm": {
"stored_tensors": {
"model.layers.18.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.18.self_attn.q_proj": {
"stored_tensors": {
"model.layers.18.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.self_attn.k_proj": {
"stored_tensors": {
"model.layers.18.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.self_attn.v_proj": {
"stored_tensors": {
"model.layers.18.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.self_attn.o_proj": {
"stored_tensors": {
"model.layers.18.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.self_attn.q_norm": {
"stored_tensors": {
"model.layers.18.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.18.self_attn.k_norm": {
"stored_tensors": {
"model.layers.18.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.18.feedforward_layernorm": {
"stored_tensors": {
"model.layers.18.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.18.mlp.up_proj": {
"stored_tensors": {
"model.layers.18.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.18.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.18.mlp.down_proj": {
"stored_tensors": {
"model.layers.18.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.18.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.18.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.attention_layernorm": {
"stored_tensors": {
"model.layers.19.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.19.self_attn.q_proj": {
"stored_tensors": {
"model.layers.19.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.self_attn.k_proj": {
"stored_tensors": {
"model.layers.19.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.self_attn.v_proj": {
"stored_tensors": {
"model.layers.19.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.self_attn.o_proj": {
"stored_tensors": {
"model.layers.19.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.self_attn.q_norm": {
"stored_tensors": {
"model.layers.19.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.19.self_attn.k_norm": {
"stored_tensors": {
"model.layers.19.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.19.feedforward_layernorm": {
"stored_tensors": {
"model.layers.19.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.19.mlp.up_proj": {
"stored_tensors": {
"model.layers.19.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.19.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.19.mlp.down_proj": {
"stored_tensors": {
"model.layers.19.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.19.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.19.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.attention_layernorm": {
"stored_tensors": {
"model.layers.20.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.20.self_attn.q_proj": {
"stored_tensors": {
"model.layers.20.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.self_attn.k_proj": {
"stored_tensors": {
"model.layers.20.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.self_attn.v_proj": {
"stored_tensors": {
"model.layers.20.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.self_attn.o_proj": {
"stored_tensors": {
"model.layers.20.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.self_attn.q_norm": {
"stored_tensors": {
"model.layers.20.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.20.self_attn.k_norm": {
"stored_tensors": {
"model.layers.20.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.20.feedforward_layernorm": {
"stored_tensors": {
"model.layers.20.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.20.mlp.up_proj": {
"stored_tensors": {
"model.layers.20.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.20.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.20.mlp.down_proj": {
"stored_tensors": {
"model.layers.20.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.20.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.20.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.attention_layernorm": {
"stored_tensors": {
"model.layers.21.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.21.self_attn.q_proj": {
"stored_tensors": {
"model.layers.21.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.self_attn.k_proj": {
"stored_tensors": {
"model.layers.21.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.self_attn.v_proj": {
"stored_tensors": {
"model.layers.21.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.self_attn.o_proj": {
"stored_tensors": {
"model.layers.21.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.self_attn.q_norm": {
"stored_tensors": {
"model.layers.21.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.21.self_attn.k_norm": {
"stored_tensors": {
"model.layers.21.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.21.feedforward_layernorm": {
"stored_tensors": {
"model.layers.21.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.21.mlp.up_proj": {
"stored_tensors": {
"model.layers.21.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.21.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.21.mlp.down_proj": {
"stored_tensors": {
"model.layers.21.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.21.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.21.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.attention_layernorm": {
"stored_tensors": {
"model.layers.22.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.22.self_attn.q_proj": {
"stored_tensors": {
"model.layers.22.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.self_attn.k_proj": {
"stored_tensors": {
"model.layers.22.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.self_attn.v_proj": {
"stored_tensors": {
"model.layers.22.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.self_attn.o_proj": {
"stored_tensors": {
"model.layers.22.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.self_attn.q_norm": {
"stored_tensors": {
"model.layers.22.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.22.self_attn.k_norm": {
"stored_tensors": {
"model.layers.22.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.22.feedforward_layernorm": {
"stored_tensors": {
"model.layers.22.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.22.mlp.up_proj": {
"stored_tensors": {
"model.layers.22.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.22.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.22.mlp.down_proj": {
"stored_tensors": {
"model.layers.22.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.22.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.22.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.attention_layernorm": {
"stored_tensors": {
"model.layers.23.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.23.self_attn.q_proj": {
"stored_tensors": {
"model.layers.23.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.self_attn.k_proj": {
"stored_tensors": {
"model.layers.23.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.self_attn.v_proj": {
"stored_tensors": {
"model.layers.23.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.self_attn.o_proj": {
"stored_tensors": {
"model.layers.23.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.self_attn.q_norm": {
"stored_tensors": {
"model.layers.23.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.23.self_attn.k_norm": {
"stored_tensors": {
"model.layers.23.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.23.feedforward_layernorm": {
"stored_tensors": {
"model.layers.23.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.23.mlp.up_proj": {
"stored_tensors": {
"model.layers.23.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.23.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.23.mlp.down_proj": {
"stored_tensors": {
"model.layers.23.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.23.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.23.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.attention_layernorm": {
"stored_tensors": {
"model.layers.24.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.24.self_attn.q_proj": {
"stored_tensors": {
"model.layers.24.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.self_attn.k_proj": {
"stored_tensors": {
"model.layers.24.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.self_attn.v_proj": {
"stored_tensors": {
"model.layers.24.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.self_attn.o_proj": {
"stored_tensors": {
"model.layers.24.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.self_attn.q_norm": {
"stored_tensors": {
"model.layers.24.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.24.self_attn.k_norm": {
"stored_tensors": {
"model.layers.24.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.24.feedforward_layernorm": {
"stored_tensors": {
"model.layers.24.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.24.mlp.up_proj": {
"stored_tensors": {
"model.layers.24.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.24.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.24.mlp.down_proj": {
"stored_tensors": {
"model.layers.24.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.24.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.24.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.attention_layernorm": {
"stored_tensors": {
"model.layers.25.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.25.self_attn.q_proj": {
"stored_tensors": {
"model.layers.25.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.self_attn.k_proj": {
"stored_tensors": {
"model.layers.25.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.self_attn.v_proj": {
"stored_tensors": {
"model.layers.25.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.self_attn.o_proj": {
"stored_tensors": {
"model.layers.25.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.self_attn.q_norm": {
"stored_tensors": {
"model.layers.25.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.25.self_attn.k_norm": {
"stored_tensors": {
"model.layers.25.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.25.feedforward_layernorm": {
"stored_tensors": {
"model.layers.25.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.25.mlp.up_proj": {
"stored_tensors": {
"model.layers.25.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.25.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.25.mlp.down_proj": {
"stored_tensors": {
"model.layers.25.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.25.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.25.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.attention_layernorm": {
"stored_tensors": {
"model.layers.26.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.26.self_attn.q_proj": {
"stored_tensors": {
"model.layers.26.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.self_attn.k_proj": {
"stored_tensors": {
"model.layers.26.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.self_attn.v_proj": {
"stored_tensors": {
"model.layers.26.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.self_attn.o_proj": {
"stored_tensors": {
"model.layers.26.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.self_attn.q_norm": {
"stored_tensors": {
"model.layers.26.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.26.self_attn.k_norm": {
"stored_tensors": {
"model.layers.26.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.26.feedforward_layernorm": {
"stored_tensors": {
"model.layers.26.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.26.mlp.up_proj": {
"stored_tensors": {
"model.layers.26.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.26.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.26.mlp.down_proj": {
"stored_tensors": {
"model.layers.26.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.26.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.26.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.attention_layernorm": {
"stored_tensors": {
"model.layers.27.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.27.self_attn.q_proj": {
"stored_tensors": {
"model.layers.27.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.self_attn.k_proj": {
"stored_tensors": {
"model.layers.27.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.self_attn.v_proj": {
"stored_tensors": {
"model.layers.27.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.self_attn.o_proj": {
"stored_tensors": {
"model.layers.27.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.self_attn.q_norm": {
"stored_tensors": {
"model.layers.27.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.27.self_attn.k_norm": {
"stored_tensors": {
"model.layers.27.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.27.feedforward_layernorm": {
"stored_tensors": {
"model.layers.27.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.27.mlp.up_proj": {
"stored_tensors": {
"model.layers.27.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.27.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.27.mlp.down_proj": {
"stored_tensors": {
"model.layers.27.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.27.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.27.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.attention_layernorm": {
"stored_tensors": {
"model.layers.28.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.28.self_attn.q_proj": {
"stored_tensors": {
"model.layers.28.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.self_attn.k_proj": {
"stored_tensors": {
"model.layers.28.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.self_attn.v_proj": {
"stored_tensors": {
"model.layers.28.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.self_attn.o_proj": {
"stored_tensors": {
"model.layers.28.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.self_attn.q_norm": {
"stored_tensors": {
"model.layers.28.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.28.self_attn.k_norm": {
"stored_tensors": {
"model.layers.28.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.28.feedforward_layernorm": {
"stored_tensors": {
"model.layers.28.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.28.mlp.up_proj": {
"stored_tensors": {
"model.layers.28.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.28.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.28.mlp.down_proj": {
"stored_tensors": {
"model.layers.28.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.28.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.28.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.attention_layernorm": {
"stored_tensors": {
"model.layers.29.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.29.self_attn.q_proj": {
"stored_tensors": {
"model.layers.29.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.self_attn.k_proj": {
"stored_tensors": {
"model.layers.29.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.self_attn.v_proj": {
"stored_tensors": {
"model.layers.29.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.self_attn.o_proj": {
"stored_tensors": {
"model.layers.29.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.self_attn.q_norm": {
"stored_tensors": {
"model.layers.29.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.29.self_attn.k_norm": {
"stored_tensors": {
"model.layers.29.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.29.feedforward_layernorm": {
"stored_tensors": {
"model.layers.29.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.29.mlp.up_proj": {
"stored_tensors": {
"model.layers.29.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.29.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.29.mlp.down_proj": {
"stored_tensors": {
"model.layers.29.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.29.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.29.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.attention_layernorm": {
"stored_tensors": {
"model.layers.30.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.30.self_attn.q_proj": {
"stored_tensors": {
"model.layers.30.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.self_attn.k_proj": {
"stored_tensors": {
"model.layers.30.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.self_attn.v_proj": {
"stored_tensors": {
"model.layers.30.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.self_attn.o_proj": {
"stored_tensors": {
"model.layers.30.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.self_attn.q_norm": {
"stored_tensors": {
"model.layers.30.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.30.self_attn.k_norm": {
"stored_tensors": {
"model.layers.30.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.30.feedforward_layernorm": {
"stored_tensors": {
"model.layers.30.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.30.mlp.up_proj": {
"stored_tensors": {
"model.layers.30.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.30.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.30.mlp.down_proj": {
"stored_tensors": {
"model.layers.30.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.30.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.30.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.attention_layernorm": {
"stored_tensors": {
"model.layers.31.attention_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.31.self_attn.q_proj": {
"stored_tensors": {
"model.layers.31.self_attn.q_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.q_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.q_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.self_attn.k_proj": {
"stored_tensors": {
"model.layers.31.self_attn.k_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.k_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.k_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.self_attn.v_proj": {
"stored_tensors": {
"model.layers.31.self_attn.v_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.v_proj.svh": {
"shape": [
1024
],
"n_bytes": 2048,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.v_proj.trellis": {
"shape": [
256,
64,
64
],
"n_bytes": 2097152,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.self_attn.o_proj": {
"stored_tensors": {
"model.layers.31.self_attn.o_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.o_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.self_attn.o_proj.trellis": {
"shape": [
256,
256,
64
],
"n_bytes": 8388608,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.self_attn.q_norm": {
"stored_tensors": {
"model.layers.31.self_attn.q_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.31.self_attn.k_norm": {
"stored_tensors": {
"model.layers.31.self_attn.k_norm.weight": {
"shape": [
128
],
"n_bytes": 256,
"dtype": "torch.float16"
}
}
},
"model.layers.31.feedforward_layernorm": {
"stored_tensors": {
"model.layers.31.feedforward_layernorm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"model.layers.31.mlp.up_proj": {
"stored_tensors": {
"model.layers.31.mlp.up_proj.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.mlp.up_proj.svh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.31.mlp.up_proj.trellis": {
"shape": [
256,
1344,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.layers.31.mlp.down_proj": {
"stored_tensors": {
"model.layers.31.mlp.down_proj.suh": {
"shape": [
21504
],
"n_bytes": 43008,
"dtype": "torch.float16"
},
"model.layers.31.mlp.down_proj.svh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"model.layers.31.mlp.down_proj.trellis": {
"shape": [
1344,
256,
64
],
"n_bytes": 44040192,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 4
},
"model.norm": {
"stored_tensors": {
"model.norm.weight": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
}
}
},
"lm_head": {
"stored_tensors": {
"lm_head.suh": {
"shape": [
4096
],
"n_bytes": 8192,
"dtype": "torch.float16"
},
"lm_head.svh": {
"shape": [
131072
],
"n_bytes": 262144,
"dtype": "torch.float16"
},
"lm_head.trellis": {
"shape": [
256,
8192,
96
],
"n_bytes": 402653184,
"dtype": "torch.int16"
}
},
"quant_format": "exl3",
"bits_per_weight": 6
}
}
}