Training in progress, epoch 1

config.json

@@ -0,0 +1,61 @@
+{
+  "_name_or_path": "answerdotai/ModernBERT-large",
+  "architectures": [
+    "ModernBertForSequenceClassification"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 50281,
+  "classifier_activation": "gelu",
+  "classifier_bias": false,
+  "classifier_dropout": 0.0,
+  "classifier_pooling": "mean",
+  "cls_token_id": 50281,
+  "decoder_bias": true,
+  "deterministic_flash_attn": false,
+  "embedding_dropout": 0.0,
+  "eos_token_id": 50282,
+  "global_attn_every_n_layers": 3,
+  "global_rope_theta": 160000.0,
+  "gradient_checkpointing": false,
+  "hidden_activation": "gelu",
+  "hidden_size": 1024,
+  "id2label": {
+    "0": 0,
+    "1": 3,
+    "2": 4,
+    "3": 1,
+    "4": 2
+  },
+  "initializer_cutoff_factor": 2.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 2624,
+  "label2id": {
+    "0": "0",
+    "1": "3",
+    "2": "4",
+    "3": "1",
+    "4": "2"
+  },
+  "layer_norm_eps": 1e-05,
+  "local_attention": 128,
+  "local_rope_theta": 10000.0,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "mlp_dropout": 0.0,
+  "model_type": "modernbert",
+  "norm_bias": false,
+  "norm_eps": 1e-05,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "pad_token_id": 50283,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "reference_compile": true,
+  "sep_token_id": 50282,
+  "sparse_pred_ignore_index": -100,
+  "sparse_prediction": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.0.dev0",
+  "vocab_size": 50368
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23f24ebf554e0e57eef59a673b5870d6292bf5fc2b5cf0f6509da46d273f7564
+size 1583363940

runs/Feb07_18-35-41_5456a55d9413/events.out.tfevents.1738953345.5456a55d9413.2447.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be49f264a1cf9b13dd837b5c84dea704064b9a0527f64ec04fb81556cdc18c86
+size 7684

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:281a7f800f44ab6fa0c2cc6284f60ffd4fd9cd031152a9d02115c6539296ef88
+size 5368