Update run_awq.py
fc3d3de verified - 1.52 kB initial commit
- 534 Bytes Update README.md
- 6.06 kB upload
- 75.9 kB upload
- 57.6 MB upload
phi3_mini_awq_4bit_no_flash_attention.pt Detected Pickle imports (24)
- "phi3_mini.modeling_phi3.Phi3DecoderLayer",
- "torch._utils._rebuild_tensor_v2",
- "phi3_mini.modeling_phi3.Phi3RotaryEmbedding",
- "torch.Size",
- "collections.OrderedDict",
- "phi3_mini.modeling_phi3.Phi3Model",
- "torch.nn.modules.activation.SiLU",
- "qlinear.QLinearPerGrp",
- "torch.bfloat16",
- "torch._utils._rebuild_parameter",
- "phi3_mini.configuration_phi3.Phi3Config",
- "torch.FloatStorage",
- "phi3_mini.modeling_phi3.Phi3MLP",
- "torch.CharStorage",
- "phi3_mini.modeling_phi3.Phi3ForCausalLM",
- "torch.nn.modules.sparse.Embedding",
- "phi3_mini.modeling_phi3.Phi3RMSNorm",
- "torch.nn.modules.container.ModuleList",
- "__builtin__.set",
- "phi3_mini.modeling_phi3.Phi3Attention",
- "torch.nn.modules.dropout.Dropout",
- "torch.nn.modules.linear.Linear",
- "torch.BFloat16Storage",
- "transformers.generation.configuration_utils.GenerationConfig"
How to fix it?
4.16 GB Upload phi3_mini_awq_4bit_no_flash_attention.pt - 8.01 kB Update run_awq.py
- 3.83 kB upload