Training in progress, epoch 1

config.json

@@ -1,14 +1,13 @@
 {
-  "_name_or_path": "bert-base-uncased",
+  "_name_or_path": "distilbert/distilbert-base-uncased",
+  "activation": "gelu",
   "architectures": [
-    "BertForTokenClassification"
+    "DistilBertForTokenClassification"
   ],
-  "attention_probs_dropout_prob": 0.1,
-  "classifier_dropout": null,
-  "gradient_checkpointing": false,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
   "id2label": {
     "0": "O",
     "1": "B-ORG",
@@ -41,7 +40,6 @@
     "28": "I-LAWYER"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
   "label2id": {
     "B-CASE_NUMBER": 19,
     "B-COURT": 15,
@@ -73,16 +71,16 @@
     "I-WITNESS": 6,
     "O": 0
   },
-  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
   "pad_token_id": 0,
-  "position_embedding_type": "absolute",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.48.2",
-  "type_vocab_size": 2,
-  "use_cache": true,
+  "transformers_version": "4.48.3",
   "vocab_size": 30522
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d317db79aad618f42fbdd32f1b8357ab5f72f9ff13a7d3fb4de3c0feb7067c3
-size 435679140
+oid sha256:798eb195be13d2a7116db8a7e7d698ffa483980039f5bcded20977243158fa5c
+size 265553068

runs/Feb16_19-32-39_943041e775be/events.out.tfevents.1739734360.943041e775be.473.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1cac85aa49d97f7771086f12f198e8af349db31b60411bbbaf644f6413c41fd
+size 6959

tokenizer_config.json

@@ -51,6 +51,6 @@
   "sep_token": "[SEP]",
   "strip_accents": null,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
+  "tokenizer_class": "DistilBertTokenizer",
   "unk_token": "[UNK]"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:372435c7e2a58d5c69f5fa011bc5593b9968a3c771cc4861341a3112c39c6a93
+oid sha256:a4322e5bdd4db85d96357c87e9c15b9e7deb27905aa4cd00c1106cabd4475b2c
 size 5368