Llama-3-Taiwan-70B-Instruct-fp8

Description

This repo contains fp8 model files for Llama-3-Taiwan-70B-Instruct.

Quantization parameter

  • activation_scheme : static
  • quant_method : fp8
  • ignored_layers : lm_head

It tooks about 8.5 hrs to quantize on H100.

Downloads last month
12
Safetensors
Model size
70.6B params
Tensor type
BF16
·
F8_E4M3
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for minyichen/Llama-3-Taiwan-70B-Instruct-fp8

Quantized
(6)
this model