Hugging Face's logo

GreenBitAI
/
Phi-3-mini-4k-instruct-layer-mix-bpw-3.0

Text Generation
Transformers
Safetensors
phi3
conversational
custom_code
text-generation-inference
Inference Endpoints
Model card Files Community
Phi-3-mini-4k-instruct-layer-mix-bpw-3.0 / quant_strategy.json
NicoNico6
update
bd89856
raw
history blame
41.3 kB
{
"measurement": {
"model.layers.0": {
"accuracy": 0.6670417785644531,
"total_bits": 357441024,
"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": {
"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": 0.7539892196655273,
"total_bits": 357441024,
"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": {
"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": 0.8989028930664062,
"total_bits": 357441024,
"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": {
"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.3": {
"accuracy": 0.8897502422332764,
"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.4": {
"accuracy": 0.9219908714294434,
"total_bits": 357441024,
"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": {
"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.5": {
"accuracy": 0.9285287857055664,
"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.6": {
"accuracy": 0.9239652156829834,
"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.7": {
"accuracy": 0.8722958564758301,
"total_bits": 357441024,
"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": {
"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.8": {
"accuracy": 0.8624000549316406,
"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": 0.8949522972106934,
"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.10": {
"accuracy": 0.8948664665222168,
"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.11": {
"accuracy": 0.8961267471313477,
"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.12": {
"accuracy": 0.8918008804321289,
"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.13": {
"accuracy": 0.9237532615661621,
"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.14": {
"accuracy": 0.941455602645874,
"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.15": {
"accuracy": 0.9418348073959351,
"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.16": {
"accuracy": 0.9384939670562744,
"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.17": {
"accuracy": 0.9345364570617676,
"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.18": {
"accuracy": 0.9283664226531982,
"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.19": {
"accuracy": 0.9368364810943604,
"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.20": {
"accuracy": 0.9356844425201416,
"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.21": {
"accuracy": 0.9530938863754272,
"total_bits": 357441024,
"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": {
"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.22": {
"accuracy": 0.9492900371551514,
"total_bits": 357441024,
"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": {
"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.23": {
"accuracy": 0.9830865263938904,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.24": {
"accuracy": 0.9818761348724365,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.25": {
"accuracy": 0.9816690683364868,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.26": {
"accuracy": 0.9815601706504822,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.27": {
"accuracy": 0.9814841151237488,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.28": {
"accuracy": 0.9800330400466919,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.29": {
"accuracy": 0.9694942235946655,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.30": {
"accuracy": 0.9758806824684143,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
},
"model.layers.31": {
"accuracy": 0.9664642810821533,
"total_bits": 456530304,
"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": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
},
"gate_up_proj": {
"group_size": {
"4": 128
},
"bits": [
4
],
"bits_prop": [
1
],
"scale_bits": 4
}
}
}
}