Angad Chandhok

[email protected]

@Adobe Bengaluru, India

Results 4 comments of


                                            Angad Chandhok

[Question] Fine-Tuning LLaVA v1.5-7B lora on Custom Dataset and RuntimeError in Model Evaluation

hi @rorubyy Were you able to figure out the reason for the hidden states being nan? I'm facing the same issue

[Error] following `model_kwargs` are not used by the model: ['image_sizes'

getting this error in MPT-7b model too

[Question] Reproduce LLaVA with Mistral backened

try using llava_llama_2 conv_template and mistralai/Mistral-7B-Instruct-v0.2. It works

🐛 [BUG] llava-v1.6-mistral-7b fail to generate right response via 'mistral_instruct' template

I think llava-v1.6-mistral-7b model uses llava_llama_2 conversation template. You can try it out!