Jayantverma2

## Pull Request Description ### Changes Made Added `conv_biomistral` template for mistral training in the ./llava/conversation.py ### Details - Updated the file to include the `conv_biomistral` template. - This template...

feature request: phi 3 in llava

1

Please integrate phi 3 with llava as it is equivalent to llama 3 on benchmarks

in order to use in vllm or lmdeploy

what is s2 in vision load tower and how it effects training

1

import os from .clip_encoder import CLIPVisionTower, CLIPVisionTowerS2 def build_vision_tower(vision_tower_cfg, **kwargs): vision_tower = getattr(vision_tower_cfg, 'mm_vision_tower', getattr(vision_tower_cfg, 'vision_tower', None)) is_absolute_path_exists = os.path.exists(vision_tower) use_s2 = getattr(vision_tower_cfg, 's2', False) if is_absolute_path_exists or vision_tower.startswith("openai") or...

Gemma template for integrating instruct tuned model

1

There are "additional_special_tokens": [ "", "" ] in gemma, should we use start_of_turn before start of text of both user ans assistantand end of turn after completing the conversation of...