Upload config.json with huggingface_hub
Browse files- config.json +8 -15
    	
        config.json
    CHANGED
    
    | @@ -1,6 +1,4 @@ | |
| 1 | 
             
            {
         | 
| 2 | 
            -
              "_attn_implementation_autoset": true,
         | 
| 3 | 
            -
              "_name_or_path": "/tmp/tmp55lzbkct",
         | 
| 4 | 
             
              "architectures": [
         | 
| 5 | 
             
                "BertModel"
         | 
| 6 | 
             
              ],
         | 
| @@ -26,35 +24,30 @@ | |
| 26 | 
             
                "auto_cast": "matmul",
         | 
| 27 | 
             
                "auto_cast_type": "bf16",
         | 
| 28 | 
             
                "compiler_type": "neuronx-cc",
         | 
| 29 | 
            -
                "compiler_version": "2. | 
| 30 | 
             
                "disable_fallback": false,
         | 
| 31 | 
             
                "disable_fast_relayout": false,
         | 
| 32 | 
             
                "dynamic_batch_size": false,
         | 
|  | |
| 33 | 
             
                "inline_weights_to_neff": true,
         | 
| 34 | 
            -
                " | 
| 35 | 
            -
             | 
| 36 | 
            -
                  "attention_mask"
         | 
| 37 | 
            -
                ],
         | 
| 38 | 
             
                "model_type": "transformer",
         | 
| 39 | 
             
                "optlevel": "2",
         | 
| 40 | 
             
                "output_attentions": false,
         | 
| 41 | 
             
                "output_hidden_states": false,
         | 
| 42 | 
            -
                " | 
| 43 | 
            -
             | 
| 44 | 
            -
             | 
| 45 | 
            -
                ],
         | 
| 46 | 
            -
                "static_batch_size": 1,
         | 
| 47 | 
            -
                "static_sequence_length": 384,
         | 
| 48 | 
             
                "tensor_parallel_size": 1
         | 
| 49 | 
             
              },
         | 
| 50 | 
             
              "num_attention_heads": 16,
         | 
| 51 | 
             
              "num_hidden_layers": 24,
         | 
| 52 | 
             
              "pad_token_id": 0,
         | 
| 53 | 
             
              "position_embedding_type": "absolute",
         | 
| 54 | 
            -
              "task": "feature-extraction",
         | 
| 55 | 
             
              "torch_dtype": "float32",
         | 
| 56 | 
             
              "torchscript": true,
         | 
| 57 | 
            -
              "transformers_version": "4. | 
| 58 | 
             
              "type_vocab_size": 2,
         | 
| 59 | 
             
              "use_cache": true,
         | 
| 60 | 
             
              "vocab_size": 30522
         | 
|  | |
| 1 | 
             
            {
         | 
|  | |
|  | |
| 2 | 
             
              "architectures": [
         | 
| 3 | 
             
                "BertModel"
         | 
| 4 | 
             
              ],
         | 
|  | |
| 24 | 
             
                "auto_cast": "matmul",
         | 
| 25 | 
             
                "auto_cast_type": "bf16",
         | 
| 26 | 
             
                "compiler_type": "neuronx-cc",
         | 
| 27 | 
            +
                "compiler_version": "2.21.18209.0+043b1bf7",
         | 
| 28 | 
             
                "disable_fallback": false,
         | 
| 29 | 
             
                "disable_fast_relayout": false,
         | 
| 30 | 
             
                "dynamic_batch_size": false,
         | 
| 31 | 
            +
                "float_dtype": "fp32",
         | 
| 32 | 
             
                "inline_weights_to_neff": true,
         | 
| 33 | 
            +
                "instance_type": "trn1",
         | 
| 34 | 
            +
                "int_dtype": "int64",
         | 
|  | |
|  | |
| 35 | 
             
                "model_type": "transformer",
         | 
| 36 | 
             
                "optlevel": "2",
         | 
| 37 | 
             
                "output_attentions": false,
         | 
| 38 | 
             
                "output_hidden_states": false,
         | 
| 39 | 
            +
                "static_batch_size": 2,
         | 
| 40 | 
            +
                "static_sequence_length": 512,
         | 
| 41 | 
            +
                "task": "feature-extraction",
         | 
|  | |
|  | |
|  | |
| 42 | 
             
                "tensor_parallel_size": 1
         | 
| 43 | 
             
              },
         | 
| 44 | 
             
              "num_attention_heads": 16,
         | 
| 45 | 
             
              "num_hidden_layers": 24,
         | 
| 46 | 
             
              "pad_token_id": 0,
         | 
| 47 | 
             
              "position_embedding_type": "absolute",
         | 
|  | |
| 48 | 
             
              "torch_dtype": "float32",
         | 
| 49 | 
             
              "torchscript": true,
         | 
| 50 | 
            +
              "transformers_version": "4.55.4",
         | 
| 51 | 
             
              "type_vocab_size": 2,
         | 
| 52 | 
             
              "use_cache": true,
         | 
| 53 | 
             
              "vocab_size": 30522
         |