Shiv Kaul

Update features.py to avoid bfloat16 unsupported error

1

Fixes https://github.com/huggingface/datasets/issues/6566 Let me know if there's any tests I need to clear.

add fp8 related changes to mistral for text-generation

1

# What does this PR do? Initial mistral fp8 change # Command Lines: 1) 128x128xbs4 ` QUANT_CONFIG=./quantization_config/maxabs_quant.json python run_generation.py --model_name_or_path mistralai/Mistral-7B-Instruct-v0.2 --attn_softmax_bf16 --use_hpu_graphs --trim_logits --use_kv_cache --reuse_cache --bf16 --batch_size 896 --fp8...

Skaulintel/ci mistral fp8

1

add CI for mistral performance

Revert "Add `--use_kv_cache` to image-to-text pipeline (#1292)"

1

This reverts commit 520c875807ef2a5b7d9bd96df335d4d0e2487f43. In order for image to text CI tests to run.