Training in progress, epoch 4, checkpoint

checkpoint-324/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:287b87f2b6473862788067656cd38225588585eeccf260ef626269210d184b90
+oid sha256:9011f0bf79b5a5fcaf9b122b9d9aa019b1ff827f49e1e9f5ab457fddd9ca3b1e
 size 592330980

checkpoint-324/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b71a4b291705809f922846a6ca1af44c2bc038ab1eb9c985afbe6ceea24c989e
+oid sha256:3fa90e5b5639a424d531ec1f5045a162e656d0d8a44c18023a6cb8b2119f25cb
 size 1014670074

checkpoint-324/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd95f0f86efa354df6a4222f1c9e390b588b8fcd802633d848de21a27eedcb23
+oid sha256:5474bfb178f1f5ac14d6be6fd7772f757be3e1728e1b25f0f54ac7eff70dac62
 size 14244

checkpoint-324/trainer_state.json

@@ -14,249 +14,249 @@
         "f1-score": 0.0,
         "precision": 0.0,
         "recall": 0.0,
-        "support": 339.0
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
         "f1-score": 0.0,
         "precision": 0.0,
         "recall": 0.0,
-        "support": 160.0
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.6898215147129765,
-        "precision": 0.6316254416961131,
-        "recall": 0.7598299681190224,
-        "support": 941.0
+        "f1-score": 0.6746310611384398,
+        "precision": 0.6075949367088608,
+        "recall": 0.7582938388625592,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5039833312905995,
-        "precision": 0.5940479630164692,
-        "recall": 0.43763303533418474,
-        "support": 4698.0
+        "f1-score": 0.4376506024096386,
+        "precision": 0.550587343690792,
+        "recall": 0.3631592101974506,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.6970899470899471,
-        "precision": 0.6303827751196173,
-        "recall": 0.7795857988165681,
-        "support": 2028.0
+        "f1-score": 0.620034542314335,
+        "precision": 0.5483008781977854,
+        "recall": 0.7133631395926477,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8719667429344848,
-        "precision": 0.8366728509585653,
-        "recall": 0.9103694233227912,
-        "support": 14861.0
+        "f1-score": 0.8620734063103671,
+        "precision": 0.8396186972154862,
+        "recall": 0.8857621736062103,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.8820403825717322,
-        "precision": 0.8925603675823639,
-        "recall": 0.8717654922180846,
-        "support": 10473.0
+        "f1-score": 0.8807729058286443,
+        "precision": 0.8630878069080317,
+        "recall": 0.8991979189247779,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.8062985074626866,
-      "eval_loss": 0.5531178712844849,
+      "eval_accuracy": 0.785908251565951,
+      "eval_loss": 0.5655931830406189,
       "eval_macro avg": {
-        "f1-score": 0.5207002740856772,
-        "precision": 0.5121841997675898,
-        "recall": 0.5370262454015216,
-        "support": 33500.0
-      },
-      "eval_runtime": 5.0026,
-      "eval_samples_per_second": 16.192,
-      "eval_steps_per_second": 2.199,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.7948191296080505,
-        "precision": 0.789408749451204,
-        "recall": 0.8062985074626866,
-        "support": 33500.0
+        "f1-score": 0.7721020318885458,
+        "precision": 0.766573839857488,
+        "recall": 0.785908251565951,
+        "support": 27619.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B-Claim": {
-        "f1-score": 0.4537205081669691,
-        "precision": 0.589622641509434,
-        "recall": 0.3687315634218289,
-        "support": 339.0
+        "f1-score": 0.36036036036036034,
+        "precision": 0.4624277456647399,
+        "recall": 0.2952029520295203,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.42794759825327516,
-        "precision": 0.7101449275362319,
-        "recall": 0.30625,
-        "support": 160.0
+        "f1-score": 0.5836909871244635,
+        "precision": 0.723404255319149,
+        "recall": 0.4892086330935252,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.7642352941176471,
-        "precision": 0.6858108108108109,
-        "recall": 0.8629117959617428,
-        "support": 941.0
+        "f1-score": 0.7396491228070177,
+        "precision": 0.6654040404040404,
+        "recall": 0.8325434439178515,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.44524236983842014,
-        "precision": 0.6338969720802202,
-        "recall": 0.3431247339293316,
-        "support": 4698.0
+        "f1-score": 0.46704216775764956,
+        "precision": 0.5973520249221184,
+        "recall": 0.3834041489627593,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.7363764327367787,
-        "precision": 0.6217317487266554,
-        "recall": 0.9028599605522682,
-        "support": 2028.0
+        "f1-score": 0.7326456649010316,
+        "precision": 0.8348157560355781,
+        "recall": 0.6527570789865872,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8781162543828609,
-        "precision": 0.8411808209047208,
-        "recall": 0.9184442500504677,
-        "support": 14861.0
+        "f1-score": 0.8679505932065659,
+        "precision": 0.8045344983428744,
+        "recall": 0.9422194777699365,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9036260459748006,
-        "precision": 0.9102800116267804,
-        "recall": 0.8970686527260575,
-        "support": 10473.0
+        "f1-score": 0.9101698606271778,
+        "precision": 0.9143513454386348,
+        "recall": 0.9060264469976155,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.820089552238806,
-      "eval_loss": 0.5071194171905518,
+      "eval_accuracy": 0.8169376154096818,
+      "eval_loss": 0.4831336438655853,
       "eval_macro avg": {
-        "f1-score": 0.6584663576386788,
-        "precision": 0.7132382761706934,
-        "recall": 0.6570558509488137,
-        "support": 33500.0
-      },
-      "eval_runtime": 4.9923,
-      "eval_samples_per_second": 16.225,
-      "eval_steps_per_second": 2.203,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8071614798424686,
-        "precision": 0.8128932519203979,
-        "recall": 0.820089552238806,
-        "support": 33500.0
+        "f1-score": 0.8047632099274016,
+        "precision": 0.8064582361050344,
+        "recall": 0.8169376154096818,
+        "support": 27619.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B-Claim": {
-        "f1-score": 0.5954198473282443,
-        "precision": 0.6170886075949367,
-        "recall": 0.5752212389380531,
-        "support": 339.0
+        "f1-score": 0.5806451612903225,
+        "precision": 0.59765625,
+        "recall": 0.5645756457564576,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.7962382445141065,
-        "precision": 0.7987421383647799,
-        "recall": 0.79375,
-        "support": 160.0
+        "f1-score": 0.7138047138047138,
+        "precision": 0.6708860759493671,
+        "recall": 0.762589928057554,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.7837431008529856,
-        "precision": 0.7423954372623575,
-        "recall": 0.8299681190223167,
-        "support": 941.0
+        "f1-score": 0.7739984882842026,
+        "precision": 0.7420289855072464,
+        "recall": 0.8088467614533965,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6291688447464715,
-        "precision": 0.6182453256626258,
-        "recall": 0.640485312899106,
-        "support": 4698.0
+        "f1-score": 0.6128912532475566,
+        "precision": 0.6068103870651641,
+        "recall": 0.6190952261934516,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8422354104846687,
-        "precision": 0.8447420634920635,
-        "recall": 0.8397435897435898,
-        "support": 2028.0
+        "f1-score": 0.8072942336126171,
+        "precision": 0.8009779951100244,
+        "recall": 0.8137108792846498,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8773398679463618,
-        "precision": 0.8876110460712072,
-        "recall": 0.8673036807751834,
-        "support": 14861.0
+        "f1-score": 0.8797559224694903,
+        "precision": 0.8951789627465303,
+        "recall": 0.8648553281580804,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9065678167474573,
-        "precision": 0.9024505629671681,
-        "recall": 0.910722811037907,
-        "support": 10473.0
+        "f1-score": 0.9226235130211126,
+        "precision": 0.912356930902925,
+        "recall": 0.933123780619987,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.843044776119403,
-      "eval_loss": 0.46074992418289185,
+      "eval_accuracy": 0.8435859372171332,
+      "eval_loss": 0.4283505082130432,
       "eval_macro avg": {
-        "f1-score": 0.7758161618028993,
-        "precision": 0.773039311630734,
-        "recall": 0.7795992503451652,
-        "support": 33500.0
-      },
-      "eval_runtime": 5.0136,
-      "eval_samples_per_second": 16.156,
-      "eval_steps_per_second": 2.194,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8436796678417321,
-        "precision": 0.8446385621767526,
-        "recall": 0.843044776119403,
-        "support": 33500.0
+        "f1-score": 0.8439412578936614,
+        "precision": 0.8447189682878564,
+        "recall": 0.8435859372171332,
+        "support": 27619.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_B-Claim": {
-        "f1-score": 0.6172839506172839,
-        "precision": 0.6472491909385113,
-        "recall": 0.5899705014749262,
-        "support": 339.0
+        "f1-score": 0.6033333333333333,
+        "precision": 0.5501519756838906,
+        "recall": 0.6678966789667896,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.7441860465116279,
-        "precision": 0.6956521739130435,
-        "recall": 0.8,
-        "support": 160.0
+        "f1-score": 0.6847457627118644,
+        "precision": 0.6474358974358975,
+        "recall": 0.7266187050359713,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.7957208354559349,
-        "precision": 0.764187866927593,
-        "recall": 0.8299681190223167,
-        "support": 941.0
+        "f1-score": 0.7818471337579618,
+        "precision": 0.7881219903691814,
+        "recall": 0.7756714060031595,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6001342582233162,
-        "precision": 0.6325471698113208,
-        "recall": 0.5708812260536399,
-        "support": 4698.0
+        "f1-score": 0.6216657593903103,
+        "precision": 0.550733024691358,
+        "recall": 0.7135716070982254,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8068448195030473,
-        "precision": 0.7689901697944593,
-        "recall": 0.8486193293885601,
-        "support": 2028.0
+        "f1-score": 0.7652218782249742,
+        "precision": 0.7960279119699409,
+        "recall": 0.7367113760556383,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.882309200297438,
-        "precision": 0.886383701188455,
-        "recall": 0.8782719870802772,
-        "support": 14861.0
+        "f1-score": 0.8739165400768475,
+        "precision": 0.8853883758826725,
+        "recall": 0.8627381792519407,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9024700070571631,
-        "precision": 0.8895381190873678,
-        "recall": 0.9157834431395016,
-        "support": 10473.0
+        "f1-score": 0.9062570845613239,
+        "precision": 0.9497505345687812,
+        "recall": 0.8665727292434424,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.8404477611940299,
-      "eval_loss": 0.5207056403160095,
+      "eval_accuracy": 0.8286324631594192,
+      "eval_loss": 0.4826153814792633,
       "eval_macro avg": {
-        "f1-score": 0.7641355882379729,
-        "precision": 0.7549354845229643,
-        "recall": 0.776213515165603,
-        "support": 33500.0
-      },
-      "eval_runtime": 4.9749,
-      "eval_samples_per_second": 16.282,
-      "eval_steps_per_second": 2.211,
+        "f1-score": 0.7481410702938023,
+        "precision": 0.7382299586573888,
+        "recall": 0.7642543830935953,
+        "support": 27619.0
+      },
+      "eval_runtime": 4.8516,
+      "eval_samples_per_second": 16.49,
+      "eval_steps_per_second": 2.061,
       "eval_weighted avg": {
-        "f1-score": 0.8386979039247872,
-        "precision": 0.8379021545974175,
-        "recall": 0.8404477611940299,
-        "support": 33500.0
+        "f1-score": 0.8345383371838407,
+        "precision": 0.8451795530099128,
+        "recall": 0.8286324631594192,
+        "support": 27619.0
       },
       "step": 324
     }
@@ -266,7 +266,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 573377029596000.0,
+  "total_flos": 575163250872000.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-324/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c84096ff478c0991daa557c55b3ae423c1b21babebadb1405f26dd17942540f
+oid sha256:96f940fbb595414fcc9335f58119ba24a5ec23e0d6daa1a329588d596e481421
 size 4856