Hugging Face's logo

GreenBitAI
/
Phi-3-mini-128k-instruct-layer-mix-bpw-2.2

Text Generation
Transformers
Safetensors
phi3
conversational
custom_code
text-generation-inference
Inference Endpoints
Model card Files Community
Phi-3-mini-128k-instruct-layer-mix-bpw-2.2 / quant_strategy.json
NicoNico6
update
7918d7a
raw
history blame contribute delete
41.2 kB
{
"measurement": {
"model.layers.0": {
"accuracy": 0.6204204559326172,
"total_bits": 289320576,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.1": {
"accuracy": 0.6884803771972656,
"total_bits": 289320576,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.2": {
"accuracy": 0.6375808715820312,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.3": {
"accuracy": 0.5973677635192871,
"total_bits": 252161664,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.4": {
"accuracy": 0.6476345062255859,
"total_bits": 252161664,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.5": {
"accuracy": 0.748924732208252,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.6": {
"accuracy": 0.7585153579711914,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.7": {
"accuracy": 0.8151102066040039,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.8": {
"accuracy": 0.8605265617370605,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.9": {
"accuracy": 0.8808994293212891,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.10": {
"accuracy": 0.8878962993621826,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.11": {
"accuracy": 0.8858823776245117,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.12": {
"accuracy": 0.8835067749023438,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.13": {
"accuracy": 0.7296466827392578,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.14": {
"accuracy": 0.9207086563110352,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.15": {
"accuracy": 0.9219722747802734,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.16": {
"accuracy": 0.9220154285430908,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.17": {
"accuracy": 0.917980432510376,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.18": {
"accuracy": 0.9091596603393555,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.19": {
"accuracy": 0.9129974842071533,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.20": {
"accuracy": 0.8978734016418457,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.21": {
"accuracy": 0.8985085487365723,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.22": {
"accuracy": 0.897059440612793,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.23": {
"accuracy": 0.8954463005065918,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.24": {
"accuracy": 0.887451171875,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.25": {
"accuracy": 0.896916389465332,
"total_bits": 252161664,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.26": {
"accuracy": 0.8984031677246094,
"total_bits": 252161664,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.27": {
"accuracy": 0.8939080238342285,
"total_bits": 233583360,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.28": {
"accuracy": 0.8929829597473145,
"total_bits": 252161664,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.29": {
"accuracy": 0.8786048889160156,
"total_bits": 283125504,
"o_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.30": {
"accuracy": 0.9041767120361328,
"total_bits": 301703808,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.31": {
"accuracy": 0.9297804832458496,
"total_bits": 400793088,
"o_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"down_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"qkv_proj": {
"group_size": {
"2": 64
},
"bits": [
2
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
}
}
}