Training in progress, epoch 1, checkpoint

checkpoint-81/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e30c0d8a28fe2d2d13d5c4ae5e3a48da9ecf5e084c4cc8b08e6b79a71de4ed20
+oid sha256:f178f656cca34783a758f900625502306f0efe537be4a2d36add6edba88bca95
 size 592330980

checkpoint-81/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6f7d9f4cbe9a45f6160570a92c1ca80f122572abccd290781818c08bce51a14
+oid sha256:3d2146852a5de0ab2f7752cd0be96f03699262953f291923bf0378dda0bec8b9
 size 1014670074

checkpoint-81/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:153cfd77c5ddcff6dcd449eda1e20921f085ad496090dd7960aa7943b2fa3326
+oid sha256:a10709715b57c56c4e889a17baa14c8e503a7581e52966680626f29eada9d5ae
 size 1064

checkpoint-81/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.32671716809272766,
+  "best_model_checkpoint": "longformer-sep_tok_full_labels/checkpoint-81",
   "epoch": 1.0,
   "eval_steps": 500,
   "global_step": 81,
@@ -11,71 +11,71 @@
     {
       "epoch": 1.0,
       "eval_B-Claim": {
-        "f1-score": 0.37160751565762,
-        "precision": 0.42788461538461536,
-        "recall": 0.3284132841328413,
-        "support": 271.0
+        "f1-score": 0.21800947867298578,
+        "precision": 0.3333333333333333,
+        "recall": 0.1619718309859155,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.3578947368421052,
-        "precision": 0.6666666666666666,
-        "recall": 0.2446043165467626,
-        "support": 139.0
+        "f1-score": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8640915593705293,
-        "precision": 0.7895424836601307,
-        "recall": 0.9541864139020537,
-        "support": 633.0
+        "f1-score": 0.8305489260143198,
+        "precision": 0.71900826446281,
+        "recall": 0.9830508474576272,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5003402749421533,
-        "precision": 0.5493126120741183,
-        "recall": 0.4593851537115721,
-        "support": 4001.0
+        "f1-score": 0.5590492696211935,
+        "precision": 0.5643589102724319,
+        "recall": 0.5538386068187393,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.7718093699515347,
-        "precision": 0.6502211636611093,
-        "recall": 0.9493293591654247,
-        "support": 2013.0
+        "f1-score": 0.7399038461538461,
+        "precision": 0.7205056179775281,
+        "recall": 0.7603754940711462,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.875016720916752,
-        "precision": 0.8846812731043188,
-        "recall": 0.865561044460127,
-        "support": 11336.0
+        "f1-score": 0.8867229093291834,
+        "precision": 0.8883965205974068,
+        "recall": 0.8850555918901243,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9992483530087988,
-        "precision": 0.9995577178239717,
-        "recall": 0.998939179632249,
-        "support": 11312.0
+        "f1-score": 0.9976548035383666,
+        "precision": 0.9961383616794018,
+        "recall": 0.9991758694577221,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8614038040733883,
-      "eval_loss": 0.31713685393333435,
+      "eval_accuracy": 0.8699050632911393,
+      "eval_loss": 0.32671716809272766,
       "eval_macro avg": {
-        "f1-score": 0.6771440758127848,
-        "precision": 0.7096952189107044,
-        "recall": 0.685774107364433,
-        "support": 29705.0
+        "f1-score": 0.6045556047614136,
+        "precision": 0.6031058583318446,
+        "recall": 0.6204954629544678,
+        "support": 31600.0
       },
-      "eval_runtime": 4.8338,
-      "eval_samples_per_second": 16.55,
-      "eval_steps_per_second": 2.069,
+      "eval_runtime": 6.234,
+      "eval_samples_per_second": 12.833,
+      "eval_steps_per_second": 1.604,
       "eval_weighted avg": {
-        "f1-score": 0.8576207231627551,
-        "precision": 0.8601529227027923,
-        "recall": 0.8614038040733883,
-        "support": 29705.0
+        "f1-score": 0.8664142595402331,
+        "precision": 0.8644597558999653,
+        "recall": 0.8699050632911393,
+        "support": 31600.0
       },
       "step": 81
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4050,
+  "max_steps": 1620,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 50,
+  "num_train_epochs": 20,
   "save_steps": 500,
   "total_flos": 143790812718000.0,
   "train_batch_size": 4,

checkpoint-81/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3aa7d1dcb1ecb96052506e9849f1d3fa19e688784f8ea37f784dfaca9546785
+oid sha256:4a6eed06a4e3317966121cd80a8275e0b82aa59ebaa0185c3b4676ff92bfed33
 size 4920