RuntimeError: FlashAttention only supports Ampere GPUs or newer.这种问题怎么解决，卡好几天了，我改代码也不行

[SongAoxiang]

root@f24a8b4b662d:/home/Telechat5/inference_telechat# python telechat_infer_demo.py Loading checkpoint shards: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 32/32 [00:12<00:00, 2.66it/s] **多轮输入演示** 提问: 你是谁？ Traceback (most recent call last): File "/home/Telechat5/inference_telechat/telechat_infer_demo.py", line 65, in main() File "/home/Telechat5/inference_telechat/telechat_infer_demo.py", line 27, in main answer, history = model.chat(tokenizer=tokenizer, question=question, history=[], generation_config=generate_config, File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 878, in chat outputs = self.generate(inputs.to(self.device), generation_config=generation_config, model_kwargs) File "/opt/conda/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, *kwargs) File "/opt/conda/lib/python3.10/site-packages/transformers/generation/utils.py", line 1522, in generate return self.greedy_search( File "/opt/conda/lib/python3.10/site-packages/transformers/generation/utils.py", line 2339, in greedy_search outputs = self( File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(input, kwargs) File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 799, in forward transformer_outputs = self.transformer( File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(*input, kwargs) File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 716, in forward outputs = block( File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(*input, *kwargs) File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 540, in forward attn_outputs = self.self_attention( File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(input, kwargs) File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 460, in forward context_layer = self.core_attention_flash(q, k, v) File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(*input, **kwargs) File "/root/.cache/huggingface/modules/transformers_modules/telechat-7B/modeling_telechat.py", line 210, in forward output = flash_attn_unpadded_func( File "/opt/conda/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 529, in flash_attn_varlen_func return FlashAttnVarlenFunc.apply( File "/opt/conda/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 288, in forward out, q, k, v, out_padded, softmax_lse, S_dmask = _flash_attn_varlen_forward( File "/opt/conda/lib/python3.10/site-packages/flash_attn/flash_attn_interface.py", line 52, in _flash_attn_varlen_forward out, q, k, v, out_padded, softmax_lse, S_dmask = flash_attn_cuda.varlen_fwd( RuntimeError: FlashAttention only supports Ampere GPUs or newer.

[oslijunw]

config里 flash_attn改为false

[cs-wilson]

这个问题我也遇到了，楼上的方案对我来说无法解决，我使用一张V100-32GB来训练，有人知道这是什么情况吗？