--batch_size 4 --num_train_epochs 15 --learning_rate 1e-5 --weight_decay 0.0

config.json

@@ -1 +1,115 @@
-{"_name_or_path": "UFNLP/gatortrons", "architectures": ["MegatronBertForTokenClassification"], "attention_probs_dropout_prob": 0.1, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 1024, "id2label": {"0": "['Living Beings', 'Virus']", "1": "['Living Beings', 'Virus']", "2": "['Living Beings', 'Bacterium']", "3": "['Living Beings', 'Bacterium']", "4": "['Anatomy', 'Anatomical Structure']", "5": "['Anatomy', 'Anatomical Structure']", "6": "['Anatomy', 'Body System']", "7": "['Anatomy', 'Body System']", "8": "['Anatomy', 'Body Substance']", "9": "['Anatomy', 'Body Substance']", "10": "['Disorders', 'Finding']", "11": "['Disorders', 'Finding']", "12": "['Disorders', 'Injury or Poisoning']", "13": "['Disorders', 'Injury or Poisoning']", "14": "['Phenomena', 'Biologic Function']", "15": "['Phenomena', 'Biologic Function']", "16": "['Procedures', 'Health Care Activity']", "17": "['Procedures', 'Health Care Activity']", "18": "['Procedures', 'Research Activity']", "19": "['Procedures', 'Research Activity']", "20": "['Devices', 'Medical Device']", "21": "['Devices', 'Medical Device']", "22": "['Concepts & Ideas', 'Spatial Concept']", "23": "['Concepts & Ideas', 'Spatial Concept']", "24": "['Occupations', 'Biomedical Occupation or Discipline']", "25": "['Occupations', 'Biomedical Occupation or Discipline']", "26": "['Organizations', 'Organization']", "27": "['Organizations', 'Organization']", "28": "['Living Beings', 'Professional or Occupational Group']", "29": "['Living Beings', 'Professional or Occupational Group']", "30": "['Living Beings', 'Population Group']", "31": "['Living Beings', 'Population Group']", "32": "['Chemicals & Drugs', 'Chemical']", "33": "['Chemicals & Drugs', 'Chemical']", "34": "['Objects', 'Food']", "35": "['Objects', 'Food']", "36": "['Concepts & Ideas', 'Intellectual Product']", "37": "['Concepts & Ideas', 'Intellectual Product']", "38": "['Physiology', 'Clinical Attribute']", "39": "['Physiology', 'Clinical Attribute']", "40": "['Living Beings', 'Eukaryote']", "41": "['Living Beings', 'Eukaryote']", "42": "None"}, "initializer_range": 0.02, "intermediate_size": 4096, "label2id": {"['Living Beings', 'Virus']": 1, "['Living Beings', 'Bacterium']": 3, "['Anatomy', 'Anatomical Structure']": 5, "['Anatomy', 'Body System']": 7, "['Anatomy', 'Body Substance']": 9, "['Disorders', 'Finding']": 11, "['Disorders', 'Injury or Poisoning']": 13, "['Phenomena', 'Biologic Function']": 15, "['Procedures', 'Health Care Activity']": 17, "['Procedures', 'Research Activity']": 19, "['Devices', 'Medical Device']": 21, "['Concepts & Ideas', 'Spatial Concept']": 23, "['Occupations', 'Biomedical Occupation or Discipline']": 25, "['Organizations', 'Organization']": 27, "['Living Beings', 'Professional or Occupational Group']": 29, "['Living Beings', 'Population Group']": 31, "['Chemicals & Drugs', 'Chemical']": 33, "['Objects', 'Food']": 35, "['Concepts & Ideas', 'Intellectual Product']": 37, "['Physiology', 'Clinical Attribute']": 39, "['Living Beings', 'Eukaryote']": 41, "None": 42}, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "megatron-bert", "num_attention_heads": 16, "num_hidden_layers": 24, "pad_token_id": 0, "position_embedding_type": "absolute", "tokenizer_type": "BertWordPieceCase", "torch_dtype": "float32", "transformers_version": "4.37.1", "type_vocab_size": 2, "use_cache": true, "vocab_size": 50176}
+{
+  "_name_or_path": "/home/ec2-user/SageMaker/LLM-NER-clinical-text/models/medmentions/gatortrons/checkpoint-5931",
+  "architectures": [
+    "MegatronBertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30",
+    "31": "LABEL_31",
+    "32": "LABEL_32",
+    "33": "LABEL_33",
+    "34": "LABEL_34",
+    "35": "LABEL_35",
+    "36": "LABEL_36",
+    "37": "LABEL_37",
+    "38": "LABEL_38",
+    "39": "LABEL_39",
+    "40": "LABEL_40",
+    "41": "LABEL_41",
+    "42": "LABEL_42"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_31": 31,
+    "LABEL_32": 32,
+    "LABEL_33": 33,
+    "LABEL_34": 34,
+    "LABEL_35": 35,
+    "LABEL_36": 36,
+    "LABEL_37": 37,
+    "LABEL_38": 38,
+    "LABEL_39": 39,
+    "LABEL_4": 4,
+    "LABEL_40": 40,
+    "LABEL_41": 41,
+    "LABEL_42": 42,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "megatron-bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tokenizer_type": "BertWordPieceCase",
+  "torch_dtype": "float32",
+  "transformers_version": "4.37.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 50176
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8505fba086bf12124180592993c06e8bcd441a9bb22d367df5efbb72d4ec1c3b
+oid sha256:a9b67a41b4af21362e62ca9506f80da0f3bea7b406a52b27df99ea72d4f17614
 size 1417093764