Training in progress, epoch 3, checkpoint

checkpoint-243/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53190b17ff35b9ad5c167544ed90396f7280afc5ddb1a6c61af3086822d56282
+oid sha256:ea4d8a7c9fe2821c281e4bde724a6657d8920522d541c7d830dd550998acd221
 size 592318676

checkpoint-243/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f41c3702d7f6a6ba317df61e02a9051a0fb6bacbce78a09c7f9cf6aed36785a
+oid sha256:e927cd86735068b468ed669860db347d0a23008ccb0d654855b74aa39ac0de93
 size 1014645498

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": 0.2252955436706543,
+  "best_model_checkpoint": "longformer-spans/checkpoint-162",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 243,
@@ -11,125 +11,125 @@
     {
       "epoch": 1.0,
       "eval_B": {
-        "f1-score": 0.8331822302810515,
-        "precision": 0.7901977644024075,
-        "recall": 0.8811121764141898,
-        "support": 1043.0
+        "f1-score": 0.816,
+        "precision": 0.7461594732991953,
+        "recall": 0.9002647837599294,
+        "support": 1133.0
       },
       "eval_I": {
-        "f1-score": 0.9471294753175105,
-        "precision": 0.9443616777446711,
-        "recall": 0.9499135446685879,
-        "support": 17350.0
+        "f1-score": 0.9321376763813793,
+        "precision": 0.9024103768767235,
+        "recall": 0.9638902525500463,
+        "support": 18333.0
       },
       "eval_O": {
-        "f1-score": 0.8992868897421833,
-        "precision": 0.9103731674811195,
-        "recall": 0.8884673748103187,
-        "support": 9226.0
+        "f1-score": 0.8560784528570645,
+        "precision": 0.931782945736434,
+        "recall": 0.7917511147142278,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.9267895289474637,
-      "eval_loss": 0.20478281378746033,
+      "eval_accuracy": 0.9035249198881844,
+      "eval_loss": 0.2619660496711731,
       "eval_macro avg": {
-        "f1-score": 0.8931995317802484,
-        "precision": 0.8816442032093993,
-        "recall": 0.9064976986310321,
-        "support": 27619.0
+        "f1-score": 0.8680720430794812,
+        "precision": 0.860117598637451,
+        "recall": 0.8853020503414012,
+        "support": 29334.0
       },
-      "eval_runtime": 4.6128,
-      "eval_samples_per_second": 17.343,
-      "eval_steps_per_second": 2.168,
+      "eval_runtime": 5.5032,
+      "eval_samples_per_second": 14.537,
+      "eval_steps_per_second": 1.817,
       "eval_weighted avg": {
-        "f1-score": 0.9268447919078652,
-        "precision": 0.9271861479533134,
-        "recall": 0.9267895289474637,
-        "support": 27619.0
+        "f1-score": 0.9020655278480035,
+        "precision": 0.9062562975065145,
+        "recall": 0.9035249198881844,
+        "support": 29334.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B": {
-        "f1-score": 0.8682547399124939,
-        "precision": 0.8806706114398422,
-        "recall": 0.8561840843720039,
-        "support": 1043.0
+        "f1-score": 0.8512256973795435,
+        "precision": 0.8167072181670721,
+        "recall": 0.8887908208296558,
+        "support": 1133.0
       },
       "eval_I": {
-        "f1-score": 0.9590983771942466,
-        "precision": 0.9592919333448654,
-        "recall": 0.9589048991354466,
-        "support": 17350.0
+        "f1-score": 0.9416781438711729,
+        "precision": 0.9152551099212274,
+        "recall": 0.9696721758577429,
+        "support": 18333.0
       },
       "eval_O": {
-        "f1-score": 0.9226525313717006,
-        "precision": 0.9208594256100194,
-        "recall": 0.9244526338608281,
-        "support": 9226.0
+        "f1-score": 0.8778173190984578,
+        "precision": 0.9380041484212952,
+        "recall": 0.8248885285772193,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.9435171439950758,
-      "eval_loss": 0.16574698686599731,
+      "eval_accuracy": 0.9178427763005387,
+      "eval_loss": 0.2252955436706543,
       "eval_macro avg": {
-        "f1-score": 0.9166685494928136,
-        "precision": 0.9202739901315756,
-        "recall": 0.9131805391227594,
-        "support": 27619.0
+        "f1-score": 0.8902403867830581,
+        "precision": 0.8899888255031981,
+        "recall": 0.8944505084215394,
+        "support": 29334.0
       },
-      "eval_runtime": 4.6238,
-      "eval_samples_per_second": 17.302,
-      "eval_steps_per_second": 2.163,
+      "eval_runtime": 5.4972,
+      "eval_samples_per_second": 14.553,
+      "eval_steps_per_second": 1.819,
       "eval_weighted avg": {
-        "f1-score": 0.9434932036816766,
-        "precision": 0.9434846863370583,
-        "recall": 0.9435171439950758,
-        "support": 27619.0
+        "f1-score": 0.9167016237671239,
+        "precision": 0.9191015935430046,
+        "recall": 0.9178427763005387,
+        "support": 29334.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B": {
-        "f1-score": 0.8791725434884815,
-        "precision": 0.8625461254612546,
-        "recall": 0.8964525407478428,
-        "support": 1043.0
+        "f1-score": 0.8531120331950207,
+        "precision": 0.8050117462803446,
+        "recall": 0.9073256840247131,
+        "support": 1133.0
       },
       "eval_I": {
-        "f1-score": 0.9588372159825352,
-        "precision": 0.9557349825345015,
-        "recall": 0.9619596541786744,
-        "support": 17350.0
+        "f1-score": 0.9469915571230095,
+        "precision": 0.9280963603037444,
+        "recall": 0.9666721213112965,
+        "support": 18333.0
       },
       "eval_O": {
-        "f1-score": 0.9204284621270084,
-        "precision": 0.9282407407407407,
-        "recall": 0.9127465857359636,
-        "support": 9226.0
+        "f1-score": 0.8903876792352629,
+        "precision": 0.9353938852934612,
+        "recall": 0.8495135792460479,
+        "support": 9868.0
       },
-      "eval_accuracy": 0.9430464535283681,
-      "eval_loss": 0.15394224226474762,
+      "eval_accuracy": 0.9249676143724006,
+      "eval_loss": 0.22786012291908264,
       "eval_macro avg": {
-        "f1-score": 0.9194794071993417,
-        "precision": 0.9155072829121655,
-        "recall": 0.9237195935541602,
-        "support": 27619.0
+        "f1-score": 0.8968304231844311,
+        "precision": 0.8895006639591835,
+        "recall": 0.9078371281940192,
+        "support": 29334.0
       },
-      "eval_runtime": 4.6233,
-      "eval_samples_per_second": 17.303,
-      "eval_steps_per_second": 2.163,
+      "eval_runtime": 5.5036,
+      "eval_samples_per_second": 14.536,
+      "eval_steps_per_second": 1.817,
       "eval_weighted avg": {
-        "f1-score": 0.9429985029052194,
-        "precision": 0.9430314866542512,
-        "recall": 0.9430464535283681,
-        "support": 27619.0
+        "f1-score": 0.9243239165827936,
+        "precision": 0.9257972230878861,
+        "recall": 0.9249676143724006,
+        "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": 431359958206800.0,
   "train_batch_size": 4,

checkpoint-243/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:821ca433db9ddb761291ab61fc06d651ad248cd9d560cdb25f7f905e5aaff14e
+oid sha256:1094850123635c13193a80398a7bb1e351dd871416d3ce265431b259bf603290
 size 4856