Upload neuronxcc-2.16.372.0+4a9b2326/MODULE_7bebb00dc9acea881b5a/config.json with huggingface_hub
Browse files
neuronxcc-2.16.372.0+4a9b2326/MODULE_7bebb00dc9acea881b5a/config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForMaskedLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"neuron": {
|
| 17 |
+
"auto_cast": null,
|
| 18 |
+
"auto_cast_type": null,
|
| 19 |
+
"compiler_type": "neuronx-cc",
|
| 20 |
+
"compiler_version": "2.16.372.0+4a9b2326",
|
| 21 |
+
"dynamic_batch_size": false,
|
| 22 |
+
"inline_weights_to_neff": true,
|
| 23 |
+
"input_names": [
|
| 24 |
+
"input_ids",
|
| 25 |
+
"attention_mask"
|
| 26 |
+
],
|
| 27 |
+
"model_type": "distilbert",
|
| 28 |
+
"optlevel": "2",
|
| 29 |
+
"output_attentions": false,
|
| 30 |
+
"output_hidden_states": false,
|
| 31 |
+
"output_names": [
|
| 32 |
+
"logits"
|
| 33 |
+
],
|
| 34 |
+
"static_batch_size": 1,
|
| 35 |
+
"static_sequence_length": 128,
|
| 36 |
+
"task": "fill-mask",
|
| 37 |
+
"tensor_parallel_size": 1
|
| 38 |
+
},
|
| 39 |
+
"pad_token_id": 0,
|
| 40 |
+
"qa_dropout": 0.1,
|
| 41 |
+
"seq_classif_dropout": 0.2,
|
| 42 |
+
"sinusoidal_pos_embds": false,
|
| 43 |
+
"tie_weights_": true,
|
| 44 |
+
"torch_dtype": "float32",
|
| 45 |
+
"torchscript": true,
|
| 46 |
+
"transformers_version": "4.51.3",
|
| 47 |
+
"vocab_size": 30522
|
| 48 |
+
}
|