Training in progress, epoch 0

config.json

@@ -1,11 +1,23 @@
 {
-  "_name_or_path": "allenai/longformer-base-4096",
+  "_name_or_path": "allenai/longformer-large-4096",
   "architectures": [
     "LongformerForSequenceClassification"
   ],
   "attention_mode": "longformer",
   "attention_probs_dropout_prob": 0.1,
   "attention_window": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
     512,
     512,
     512,
@@ -24,14 +36,14 @@
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "NOT_REJECTED",
     "1": "REJECTED"
   },
   "ignore_attention_mask": false,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "NOT_REJECTED": 0,
     "REJECTED": 1
@@ -39,8 +51,8 @@
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 4098,
   "model_type": "longformer",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "onnx_export": false,
   "pad_token_id": 1,
   "problem_type": "single_label_classification",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9f74542581bbccf1194256a309f56f0b277d6ec9ccc051008deac0a7767d919
-size 594739281
+oid sha256:6d5bb47fee091229e8e149568c13adcaf73a02567f6cd9f2ae2b52eba9fa7a5d
+size 1738600369

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b37d494b740eb8716d42ac6cb1ff9778e0ad55c2b3c076e78b2d2b017f1bfcc6
+oid sha256:7455f37f756b8941e475a8752b273b1b7d76b12b9d6585e25abd9ca1fc85b31c
 size 3963