Training in progress, epoch 3, checkpoint

checkpoint-243/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41943dc9660f79944ddc19eaaf6f8e3ed7fb4459ccb911a39775840b6172f7de
+oid sha256:e0964f529c878d0a983e33430fce3d86f62b4fad606a1f538a0ba285aaf6954d
 size 592330980

checkpoint-243/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a050518586002d9e18ed8933db338851a981993133d2f1d7107a4c6a7c4b2f76
+oid sha256:6cf401be720085aa3cf139c37c2383f5ec5dd2bad2640c09225eb609824490c7
 size 1014670074

checkpoint-243/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e7403b5a2b8007c90df75059abe86b7e67f96587685c6c4fcbd47bcd5bb1b47
+oid sha256:9e984b98ab63d56b5554ebc9d5cbb96c54b5395d60b231b99094a49c73bd68ca
 size 1064

checkpoint-243/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.5138605833053589,
+  "best_model_checkpoint": "longformer-full_labels/checkpoint-162",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 243,
@@ -11,197 +11,197 @@
     {
       "epoch": 1.0,
       "eval_B-Claim": {
-        "f1-score": 0.0,
-        "precision": 0.0,
-        "recall": 0.0,
-        "support": 271.0
+        "f1-score": 0.006993006993006993,
+        "precision": 0.5,
+        "recall": 0.0035211267605633804,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
         "f1-score": 0.0,
         "precision": 0.0,
         "recall": 0.0,
-        "support": 139.0
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.6746310611384398,
-        "precision": 0.6075949367088608,
-        "recall": 0.7582938388625592,
-        "support": 633.0
+        "f1-score": 0.6336375488917861,
+        "precision": 0.5883777239709443,
+        "recall": 0.6864406779661016,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.4376506024096386,
-        "precision": 0.550587343690792,
-        "recall": 0.3631592101974506,
-        "support": 4001.0
+        "f1-score": 0.37228850325379614,
+        "precision": 0.41618672324946954,
+        "recall": 0.3367672308069659,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.620034542314335,
-        "precision": 0.5483008781977854,
-        "recall": 0.7133631395926477,
-        "support": 2013.0
+        "f1-score": 0.49153013051930017,
+        "precision": 0.5611921369689283,
+        "recall": 0.43725296442687744,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8620734063103671,
-        "precision": 0.8396186972154862,
-        "recall": 0.8857621736062103,
-        "support": 11336.0
+        "f1-score": 0.846019669697532,
+        "precision": 0.7746890504995582,
+        "recall": 0.9318181818181818,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.8807729058286443,
-        "precision": 0.8630878069080317,
-        "recall": 0.8991979189247779,
-        "support": 9226.0
+        "f1-score": 0.8650519031141869,
+        "precision": 0.9111808904340025,
+        "recall": 0.8233684637211187,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.785908251565951,
-      "eval_loss": 0.5655931830406189,
+      "eval_accuracy": 0.7591191109292971,
+      "eval_loss": 0.6442636251449585,
       "eval_macro avg": {
-        "f1-score": 0.49645178828591785,
-        "precision": 0.48702709467442235,
-        "recall": 0.5171108973119495,
-        "support": 27619.0
-      },
-      "eval_runtime": 4.8252,
-      "eval_samples_per_second": 16.58,
-      "eval_steps_per_second": 2.072,
+        "f1-score": 0.4593601089242298,
+        "precision": 0.535946646446129,
+        "recall": 0.4598812350714013,
+        "support": 29334.0
+      },
+      "eval_runtime": 6.1153,
+      "eval_samples_per_second": 13.082,
+      "eval_steps_per_second": 1.635,
       "eval_weighted avg": {
-        "f1-score": 0.7721020318885458,
-        "precision": 0.766573839857488,
-        "recall": 0.785908251565951,
-        "support": 27619.0
+        "f1-score": 0.7448054609057474,
+        "precision": 0.7451676238152983,
+        "recall": 0.7591191109292971,
+        "support": 29334.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B-Claim": {
-        "f1-score": 0.36036036036036034,
-        "precision": 0.4624277456647399,
-        "recall": 0.2952029520295203,
-        "support": 271.0
+        "f1-score": 0.47385620915032683,
+        "precision": 0.4420731707317073,
+        "recall": 0.5105633802816901,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.5836909871244635,
-        "precision": 0.723404255319149,
-        "recall": 0.4892086330935252,
-        "support": 139.0
+        "f1-score": 0.5267489711934156,
+        "precision": 0.6274509803921569,
+        "recall": 0.45390070921985815,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.7396491228070177,
-        "precision": 0.6654040404040404,
-        "recall": 0.8325434439178515,
-        "support": 633.0
+        "f1-score": 0.7686170212765958,
+        "precision": 0.7261306532663316,
+        "recall": 0.8163841807909604,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.46704216775764956,
-        "precision": 0.5973520249221184,
-        "recall": 0.3834041489627593,
-        "support": 4001.0
+        "f1-score": 0.5068787075602675,
+        "precision": 0.5221008840353614,
+        "recall": 0.49251900907530044,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.7326456649010316,
-        "precision": 0.8348157560355781,
-        "recall": 0.6527570789865872,
-        "support": 2013.0
+        "f1-score": 0.6644462947543713,
+        "precision": 0.5741007194244604,
+        "recall": 0.7885375494071146,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8679505932065659,
-        "precision": 0.8045344983428744,
-        "recall": 0.9422194777699365,
-        "support": 11336.0
+        "f1-score": 0.8735948241002829,
+        "precision": 0.8642982877260361,
+        "recall": 0.8830935251798561,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9101698606271778,
-        "precision": 0.9143513454386348,
-        "recall": 0.9060264469976155,
-        "support": 9226.0
+        "f1-score": 0.8825588796944621,
+        "precision": 0.925979519145147,
+        "recall": 0.8430279691933522,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.8169376154096818,
-      "eval_loss": 0.4831336438655853,
+      "eval_accuracy": 0.8015272380173177,
+      "eval_loss": 0.5138605833053589,
       "eval_macro avg": {
-        "f1-score": 0.6659298223977524,
-        "precision": 0.7146128094467336,
-        "recall": 0.6430517402511137,
-        "support": 27619.0
-      },
-      "eval_runtime": 4.8125,
-      "eval_samples_per_second": 16.623,
-      "eval_steps_per_second": 2.078,
+        "f1-score": 0.6709572725328175,
+        "precision": 0.668876316388743,
+        "recall": 0.6840037604497332,
+        "support": 29334.0
+      },
+      "eval_runtime": 6.1482,
+      "eval_samples_per_second": 13.012,
+      "eval_steps_per_second": 1.627,
       "eval_weighted avg": {
-        "f1-score": 0.8047632099274016,
-        "precision": 0.8064582361050344,
-        "recall": 0.8169376154096818,
-        "support": 27619.0
+        "f1-score": 0.8031401896750007,
+        "precision": 0.8089032379475054,
+        "recall": 0.8015272380173177,
+        "support": 29334.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B-Claim": {
-        "f1-score": 0.5806451612903225,
-        "precision": 0.59765625,
-        "recall": 0.5645756457564576,
-        "support": 271.0
+        "f1-score": 0.4233576642335767,
+        "precision": 0.4393939393939394,
+        "recall": 0.4084507042253521,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.7138047138047138,
-        "precision": 0.6708860759493671,
-        "recall": 0.762589928057554,
-        "support": 139.0
+        "f1-score": 0.6294820717131473,
+        "precision": 0.7181818181818181,
+        "recall": 0.5602836879432624,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.7739984882842026,
-        "precision": 0.7420289855072464,
-        "recall": 0.8088467614533965,
-        "support": 633.0
+        "f1-score": 0.7608562691131499,
+        "precision": 0.6709816612729234,
+        "recall": 0.8785310734463276,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6128912532475566,
-        "precision": 0.6068103870651641,
-        "recall": 0.6190952261934516,
-        "support": 4001.0
+        "f1-score": 0.46000546000546,
+        "precision": 0.5186211141889813,
+        "recall": 0.41329408879077756,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8072942336126171,
-        "precision": 0.8009779951100244,
-        "recall": 0.8137108792846498,
-        "support": 2013.0
+        "f1-score": 0.6850351161534306,
+        "precision": 0.7556615017878426,
+        "recall": 0.6264822134387352,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8797559224694903,
-        "precision": 0.8951789627465303,
-        "recall": 0.8648553281580804,
-        "support": 11336.0
+        "f1-score": 0.8689253296477533,
+        "precision": 0.7999862438957287,
+        "recall": 0.9508665794637018,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9226235130211126,
-        "precision": 0.912356930902925,
-        "recall": 0.933123780619987,
-        "support": 9226.0
+        "f1-score": 0.8740981828044481,
+        "precision": 0.9404692424419283,
+        "recall": 0.8164775030401297,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.8435859372171332,
-      "eval_loss": 0.4283505082130432,
+      "eval_accuracy": 0.7996863707643008,
+      "eval_loss": 0.5769864320755005,
       "eval_macro avg": {
-        "f1-score": 0.7558590408185736,
-        "precision": 0.7465565124687511,
-        "recall": 0.7666853642176539,
-        "support": 27619.0
-      },
-      "eval_runtime": 4.8554,
-      "eval_samples_per_second": 16.476,
-      "eval_steps_per_second": 2.06,
+        "f1-score": 0.6716800133815666,
+        "precision": 0.6918993601661659,
+        "recall": 0.6649122643354695,
+        "support": 29334.0
+      },
+      "eval_runtime": 6.0693,
+      "eval_samples_per_second": 13.181,
+      "eval_steps_per_second": 1.648,
       "eval_weighted avg": {
-        "f1-score": 0.8439412578936614,
-        "precision": 0.8447189682878564,
-        "recall": 0.8435859372171332,
-        "support": 27619.0
+        "f1-score": 0.7930703491848509,
+        "precision": 0.7980829724295806,
+        "recall": 0.7996863707643008,
+        "support": 29334.0
       },
       "step": 243
     }
   ],
   "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": 431372438154000.0,
   "train_batch_size": 4,

checkpoint-243/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96f940fbb595414fcc9335f58119ba24a5ec23e0d6daa1a329588d596e481421
-size 4856
+oid sha256:5653e8d83a92b07eb1df6cde8969d97342166ccb896f2bdca8c3721b3a61a616
+size 4920