runtime error

available: 22.3.1 -> 24.1.2 [notice] To update, run: /usr/local/bin/python -m pip install --upgrade pip Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. Downloading shards: 0%| | 0/4 [00:00<?, ?it/s] Downloading shards: 25%|██▌ | 1/4 [00:12<00:36, 12.16s/it] Downloading shards: 50%|█████ | 2/4 [00:25<00:25, 12.57s/it] Downloading shards: 75%|███████▌ | 3/4 [00:38<00:12, 12.94s/it] Downloading shards: 100%|██████████| 4/4 [00:42<00:00, 9.52s/it] Downloading shards: 100%|██████████| 4/4 [00:42<00:00, 10.67s/it] The model was loaded with use_flash_attention_2=True, which is deprecated and may be removed in a future release. Please use `attn_implementation="flash_attention_2"` instead. Traceback (most recent call last): File "/home/user/app/app.py", line 4, in <module> from chatbot import chatbot, model_inference, EXAMPLES File "/home/user/app/chatbot.py", line 33, in <module> model = LlavaForConditionalGeneration.from_pretrained(model_id, torch_dtype=torch.float16, use_flash_attention_2=True, low_cpu_mem_usage=True) File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 3704, in from_pretrained config = cls._autoset_attn_implementation( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 1481, in _autoset_attn_implementation cls._check_and_enable_flash_attn_2( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 1592, in _check_and_enable_flash_attn_2 raise ImportError(f"{preface} Flash Attention 2 is not available. {install_message}") ImportError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: Flash Attention 2 is not available. Please refer to the documentation of https://huggingface.co/docs/transformers/perf_infer_gpu_one#flashattention-2 to install Flash Attention 2.

Container logs: