Update README.md
f950c8a verified - 1.68 kB Rename phi4-mini-8dq4w.pte to phi4-mini-8da4w.pte
- 1.04 kB Create LICENSE
- 10.3 kB Update README.md
- 249 Bytes Upload tokenizer
- 423 Bytes Upload tokenizer
- 4.85 kB Upload Phi3ForCausalLM
- 174 Bytes Upload Phi3ForCausalLM
- 2.42 MB Upload tokenizer
- 3.02 GB Rename phi4-mini-8dq4w.pte to phi4-mini-8da4w.pte
pytorch_model.bin Detected Pickle imports (17)
- "torchao.dtypes.uintx.q_dq_layout.QDQTensorImpl",
- "torch._utils._rebuild_tensor_v2",
- "torch.CharStorage",
- "torchao.quantization.quant_api._int8_asymm_per_token_quant",
- "torch._tensor._rebuild_from_type_v2",
- "torchao.quantization.quant_primitives.ZeroPointDomain",
- "torchao.quantization.linear_activation_quantized_tensor.LinearActivationQuantizedTensor",
- "torchao.dtypes.uintx.q_dq_layout.QDQLayout",
- "torch.int8",
- "torch.device",
- "torch._utils._rebuild_wrapper_subclass",
- "torch.BFloat16Storage",
- "torch.serialization._get_layout",
- "torch.FloatStorage",
- "torch.float32",
- "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
- "collections.OrderedDict"
How to fix it?
4.81 GB Upload Phi3ForCausalLM pytorch_model_converted.bin Detected Pickle imports (17)
- "torchao.quantization.quant_api._int8_asymm_per_token_quant",
- "collections.OrderedDict",
- "torch._tensor._rebuild_from_type_v2",
- "torch.float32",
- "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
- "torch.FloatStorage",
- "torchao.dtypes.uintx.q_dq_layout.QDQLayout",
- "torch._utils._rebuild_tensor_v2",
- "torchao.quantization.linear_activation_quantized_tensor.LinearActivationQuantizedTensor",
- "torch.device",
- "torch.CharStorage",
- "torch.BFloat16Storage",
- "torch.int8",
- "torch._utils._rebuild_wrapper_subclass",
- "torch.serialization._get_layout",
- "torchao.quantization.quant_primitives.ZeroPointDomain",
- "torchao.dtypes.uintx.q_dq_layout.QDQTensorImpl"
How to fix it?
4.81 GB Rename phi4-mini-8da4w-converted.bin to pytorch_model_converted.bin - 587 Bytes Upload tokenizer
- 15.5 MB Upload tokenizer
- 2.52 kB Upload tokenizer
- 3.91 MB Upload tokenizer
