runtime error

le, you can pin a revision. Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. Downloading shards: 0%| | 0/7 [00:00<?, ?it/s] Downloading shards: 14%|█▍ | 1/7 [00:15<01:32, 15.34s/it] Downloading shards: 29%|██▊ | 2/7 [01:01<02:47, 33.55s/it] Downloading shards: 43%|████▎ | 3/7 [01:39<02:21, 35.36s/it] Downloading shards: 57%|█████▋ | 4/7 [02:17<01:49, 36.52s/it] Downloading shards: 71%|███████▏ | 5/7 [02:55<01:14, 37.16s/it] Downloading shards: 86%|████████▌ | 6/7 [03:32<00:37, 37.07s/it] Downloading shards: 100%|██████████| 7/7 [04:05<00:00, 35.55s/it] Downloading shards: 100%|██████████| 7/7 [04:05<00:00, 35.01s/it] Traceback (most recent call last): File "/home/user/app/app.py", line 107, in <module> "idefics2-8b-chatty": Idefics2ForConditionalGeneration.from_pretrained( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 3544, in from_pretrained config = cls._autoset_attn_implementation( File "/usr/local/lib/python3.10/site-packages/transformers/models/idefics2/modeling_idefics2.py", line 1385, in _autoset_attn_implementation config = super()._autoset_attn_implementation( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 1454, in _autoset_attn_implementation cls._check_and_enable_flash_attn_2( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 1555, 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: