Upload config.json with huggingface_hub

config.json

@@ -1,14 +1,12 @@
 {
-  "_name_or_path": "dslim/bert-base-NER",
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "/tmp/tmp85t7b0_b",
   "_num_labels": 9,
   "architectures": [
     "BertForTokenClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "auto_cast": null,
-  "auto_cast_type": null,
   "classifier_dropout": null,
-  "dynamic_batch_size": true,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -24,11 +22,6 @@
     "8": "I-LOC"
   },
   "initializer_range": 0.02,
-  "input_names": [
-    "input_ids",
-    "attention_mask",
-    "token_type_ids"
-  ],
   "intermediate_size": 3072,
   "label2id": {
     "B-LOC": 7,
@@ -44,18 +37,40 @@
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "neuron_batch_size": 1,
-  "neuron_sequence_length": 128,
+  "neuron": {
+    "auto_cast": "matmul",
+    "auto_cast_type": "bf16",
+    "compiler_type": "neuronx-cc",
+    "compiler_version": "2.16.372.0+4a9b2326",
+    "disable_fallback": false,
+    "disable_fast_relayout": false,
+    "dynamic_batch_size": false,
+    "inline_weights_to_neff": true,
+    "input_names": [
+      "input_ids",
+      "attention_mask",
+      "token_type_ids"
+    ],
+    "model_type": "bert",
+    "optlevel": "2",
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_names": [
+      "logits"
+    ],
+    "static_batch_size": 1,
+    "static_sequence_length": 128,
+    "tensor_parallel_size": 1
+  },
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
-  "output_names": [
-    "logits"
-  ],
   "output_past": true,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "task": "token-classification",
+  "torch_dtype": "float32",
   "torchscript": true,
-  "transformers_version": "4.29.2",
+  "transformers_version": "4.49.0",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 28996