Hugging Face's logo

nm-testing
/
Meta-Llama-3-8B-Instruct-fp8-hf_compat

Safetensors
llama
compressed-tensors
Model card Files Community
Meta-Llama-3-8B-Instruct-fp8-hf_compat
File size: 124 Bytes 
14a1fe3
 
 
 
 
 
 
1
2
3
4
5
6
7
 
quant_stage:
  quant_modifiers:
    QuantizationModifier:
      ignore: [lm_head]
      targets: [Linear]
      scheme: FP8