TorchAO Testing
Update README.md
1dc1fef verified - 1.52 kB initial commit
- 1.74 kB Update README.md
- 2.07 kB Upload OPTForCausalLM
- 132 Bytes Upload OPTForCausalLM
- 456 kB Upload tokenizer
- 2.67 kB Upload model_output.pt with huggingface_hub
- 1.26 kB Upload model_prompt.pt with huggingface_hub
pytorch_model.bin Detected Pickle imports (20)
- "torch._utils._rebuild_wrapper_subclass",
- "torch._tensor._rebuild_from_type_v2",
- "torch.float8_e4m3fn",
- "collections.OrderedDict",
- "torchao.dtypes.floatx.float8_layout.Float8AQTTensorImpl",
- "torchao.quantization.quant_api._input_activation_quant_func_fp8",
- "torch.serialization._get_layout",
- "torchao.quantization.linear_activation_quantized_tensor.LinearActivationQuantizedTensor",
- "torch.BFloat16Storage",
- "torch.bfloat16",
- "torch.FloatStorage",
- "torchao.quantization.granularity.PerRow",
- "torch._utils._rebuild_tensor_v2",
- "torch._utils._rebuild_tensor_v3",
- "torchao.float8.inference.Float8MMConfig",
- "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
- "torchao.dtypes.floatx.float8_layout.Float8Layout",
- "torch.storage.UntypedStorage",
- "torchao.quantization.quant_primitives.ZeroPointDomain",
- "torch.device"
How to fix it?
166 MB Upload OPTForCausalLM - 548 Bytes Upload tokenizer
- 3.56 MB Upload tokenizer
- 700 Bytes Upload tokenizer
- 798 kB Upload tokenizer
