Training in progress, epoch 5, checkpoint

checkpoint-405/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e69b36a64f36637ae6283e39b849bc0fa9d7d19e0d18bfb409fca1c041fd2a77
+oid sha256:fe5af4c5796f104388c3e333869ad6430486b3377ad6300fded70af9c04f789d
 size 592324828

checkpoint-405/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21a5136dd5226be50aa7dd904444e30396ea279c9375c1ee5d6dd72199fa7954
+oid sha256:6f2bf6d56e8265bab051efecbff6c171f21022a378ec5c93b88bd03ee72065e1
 size 1014657786

checkpoint-405/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf337692f531efc3a23b910237391cc3da6dc28e56ab2fbd5db20c81d4fb98bd
+oid sha256:d68d97adc90f9638b68f64674fd8773cc664d1043303df18c9b1b59a1cb82038
 size 14244

checkpoint-405/trainer_state.json

@@ -11,225 +11,225 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.5385059461955894,
-        "precision": 0.6267132491265789,
-        "recall": 0.4720647773279352,
-        "support": 4940.0
+        "f1-score": 0.5254768748366867,
+        "precision": 0.5768789443488239,
+        "recall": 0.4824856046065259,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7814784727863526,
-        "precision": 0.9588039867109635,
-        "recall": 0.6595063985374772,
-        "support": 2188.0
+        "f1-score": 0.7984361424847959,
+        "precision": 0.7495921696574225,
+        "recall": 0.854089219330855,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9938608715302625,
-        "precision": 0.9996969237763298,
-        "recall": 0.9880925634688834,
-        "support": 13353.0
+        "f1-score": 0.9990277532260916,
+        "precision": 0.9988511841640155,
+        "recall": 0.9992043847241867,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.8968837690149166,
-        "precision": 0.8455112497215416,
-        "recall": 0.954902824076986,
-        "support": 15899.0
+        "f1-score": 0.8854183656825966,
+        "precision": 0.8719781543651113,
+        "recall": 0.8992793837488611,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8837548103353491,
-      "eval_loss": 0.2860471308231354,
+      "eval_accuracy": 0.8755765022723447,
+      "eval_loss": 0.28303390741348267,
       "eval_macro avg": {
-        "f1-score": 0.8026822648817803,
-        "precision": 0.8576813523338533,
-        "recall": 0.7686416408528204,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9572,
-      "eval_samples_per_second": 16.34,
-      "eval_steps_per_second": 2.219,
+        "f1-score": 0.8020897840575427,
+        "precision": 0.7993251131338432,
+        "recall": 0.8087646481026072,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8106,
+      "eval_samples_per_second": 16.63,
+      "eval_steps_per_second": 2.079,
       "eval_weighted avg": {
-        "f1-score": 0.8768739289107891,
-        "precision": 0.8792073379911217,
-        "recall": 0.8837548103353491,
-        "support": 36380.0
+        "f1-score": 0.8718761173649596,
+        "precision": 0.8700202202343714,
+        "recall": 0.8755765022723447,
+        "support": 29705.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.6469459868753155,
-        "precision": 0.6453172205438067,
-        "recall": 0.648582995951417,
-        "support": 4940.0
+        "f1-score": 0.5578782336033447,
+        "precision": 0.6124497991967871,
+        "recall": 0.5122360844529751,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8435970110796187,
-        "precision": 0.9669226225634967,
-        "recall": 0.7481718464351006,
-        "support": 2188.0
+        "f1-score": 0.784421283598464,
+        "precision": 0.9571619812583668,
+        "recall": 0.6644981412639405,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.998877245508982,
-        "precision": 0.9983541557567143,
-        "recall": 0.9994008836965476,
-        "support": 13353.0
+        "f1-score": 0.9998674150351351,
+        "precision": 0.9997348652231551,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9099026477336144,
-        "precision": 0.8972179761540813,
-        "recall": 0.922951129001824,
-        "support": 15899.0
+        "f1-score": 0.8977749872463995,
+        "precision": 0.8530201342281879,
+        "recall": 0.9474861260664292,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.903243540406817,
-      "eval_loss": 0.24885737895965576,
+      "eval_accuracy": 0.8859114627167144,
+      "eval_loss": 0.27363118529319763,
       "eval_macro avg": {
-        "f1-score": 0.8498307227993828,
-        "precision": 0.8769529937545246,
-        "recall": 0.8297767137712222,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0097,
-      "eval_samples_per_second": 16.169,
-      "eval_steps_per_second": 2.196,
+        "f1-score": 0.8099854798708358,
+        "precision": 0.8555916949766242,
+        "recall": 0.7810550879458362,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7727,
+      "eval_samples_per_second": 16.762,
+      "eval_steps_per_second": 2.095,
       "eval_weighted avg": {
-        "f1-score": 0.9028656814459712,
-        "precision": 0.9043261520711511,
-        "recall": 0.903243540406817,
-        "support": 36380.0
+        "f1-score": 0.8807489883812781,
+        "precision": 0.8826802296805741,
+        "recall": 0.8859114627167144,
+        "support": 29705.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.7040630102775911,
-        "precision": 0.6698958142935478,
-        "recall": 0.7419028340080972,
-        "support": 4940.0
+        "f1-score": 0.64760736196319,
+        "precision": 0.6627322953289804,
+        "recall": 0.6331573896353166,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8888411094388303,
-        "precision": 0.8392204628501827,
-        "recall": 0.9446983546617916,
-        "support": 2188.0
+        "f1-score": 0.9016018306636155,
+        "precision": 0.8881875563570785,
+        "recall": 0.9154275092936803,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.998989104796136,
-        "precision": 0.9988769092542678,
-        "recall": 0.9991013255448213,
-        "support": 13353.0
+        "f1-score": 0.9996904982977407,
+        "precision": 1.0,
+        "recall": 0.9993811881188119,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.914582593823615,
-        "precision": 0.9390987408880053,
-        "recall": 0.8913139191144097,
-        "support": 15899.0
+        "f1-score": 0.8982820417748115,
+        "precision": 0.8936065573770492,
+        "recall": 0.9030067091857865,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9137987905442551,
-      "eval_loss": 0.24298855662345886,
+      "eval_accuracy": 0.9027436458508669,
+      "eval_loss": 0.22585685551166534,
       "eval_macro avg": {
-        "f1-score": 0.8766189545840432,
-        "precision": 0.8617729818215009,
-        "recall": 0.8942541083322799,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9876,
-      "eval_samples_per_second": 16.24,
-      "eval_steps_per_second": 2.205,
+        "f1-score": 0.8617954331748394,
+        "precision": 0.861131602265777,
+        "recall": 0.8627431990583988,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8003,
+      "eval_samples_per_second": 16.666,
+      "eval_steps_per_second": 2.083,
       "eval_weighted avg": {
-        "f1-score": 0.9154289662937856,
-        "precision": 0.9184781183611038,
-        "recall": 0.9137987905442551,
-        "support": 36380.0
+        "f1-score": 0.9019670975035187,
+        "precision": 0.9013351218793044,
+        "recall": 0.9027436458508669,
+        "support": 29705.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6981611893583725,
-        "precision": 0.6754352763058289,
-        "recall": 0.7224696356275304,
-        "support": 4940.0
+        "f1-score": 0.6477602966229081,
+        "precision": 0.5561865427637239,
+        "recall": 0.7754318618042226,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.901231802911534,
-        "precision": 0.8836187966622749,
-        "recall": 0.9195612431444241,
-        "support": 2188.0
+        "f1-score": 0.8984409799554566,
+        "precision": 0.8627031650983746,
+        "recall": 0.9372676579925651,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9989130842172332,
-        "precision": 0.9998499399759904,
-        "recall": 0.9979779824758481,
-        "support": 13353.0
+        "f1-score": 1.0,
+        "precision": 1.0,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9146224912392481,
-        "precision": 0.9266670970240785,
-        "recall": 0.9028869740235235,
-        "support": 15899.0
+        "f1-score": 0.861495720750997,
+        "precision": 0.9384029675907849,
+        "recall": 0.7962395427814131,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9142935678944475,
-      "eval_loss": 0.23980534076690674,
+      "eval_accuracy": 0.8811311227066151,
+      "eval_loss": 0.32341843843460083,
       "eval_macro avg": {
-        "f1-score": 0.8782321419315969,
-        "precision": 0.8713927774920431,
-        "recall": 0.8857239588178316,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9726,
-      "eval_samples_per_second": 16.289,
-      "eval_steps_per_second": 2.212,
+        "f1-score": 0.8519242493323405,
+        "precision": 0.8393231688632209,
+        "recall": 0.8772347656445502,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7708,
+      "eval_samples_per_second": 16.769,
+      "eval_steps_per_second": 2.096,
       "eval_weighted avg": {
-        "f1-score": 0.9153623106642749,
-        "precision": 0.9168247558035478,
-        "recall": 0.9142935678944475,
-        "support": 36380.0
+        "f1-score": 0.8869263673393438,
+        "precision": 0.9027457246003854,
+        "recall": 0.8811311227066151,
+        "support": 29705.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_Claim": {
-        "f1-score": 0.6650318783717509,
-        "precision": 0.6450999048525214,
-        "recall": 0.6862348178137652,
-        "support": 4940.0
+        "f1-score": 0.6188788236838649,
+        "precision": 0.6833864888373442,
+        "recall": 0.5654990403071017,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9172031076581575,
-        "precision": 0.8916702632714717,
-        "recall": 0.9442413162705667,
-        "support": 2188.0
+        "f1-score": 0.909733679000707,
+        "precision": 0.9230033476805356,
+        "recall": 0.8968401486988847,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9982383147794145,
-        "precision": 0.9992495872730002,
-        "recall": 0.9972290870965326,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.8998438545616774,
-        "precision": 0.9119622787753521,
-        "recall": 0.8880432731618341,
-        "support": 15899.0
+        "f1-score": 0.8978936810431294,
+        "precision": 0.8706816059757236,
+        "recall": 0.9268615919821088,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9040956569543706,
-      "eval_loss": 0.2932502031326294,
+      "eval_accuracy": 0.9018347079616226,
+      "eval_loss": 0.3072384297847748,
       "eval_macro avg": {
-        "f1-score": 0.87007928884275,
-        "precision": 0.8619955085430864,
-        "recall": 0.8789371235856747,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9595,
-      "eval_samples_per_second": 16.332,
-      "eval_steps_per_second": 2.218,
+        "f1-score": 0.8566154962081685,
+        "precision": 0.8692457621526151,
+        "recall": 0.8473001952470238,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7922,
+      "eval_samples_per_second": 16.694,
+      "eval_steps_per_second": 2.087,
       "eval_weighted avg": {
-        "f1-score": 0.9051179642561333,
-        "precision": 0.9065430476942037,
-        "recall": 0.9040956569543706,
-        "support": 36380.0
+        "f1-score": 0.8984684143294738,
+        "precision": 0.8974043833368577,
+        "recall": 0.9018347079616226,
+        "support": 29705.0
       },
       "step": 405
     }
@@ -239,7 +239,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 716710919337000.0,
+  "total_flos": 718943663634000.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-405/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95e8ce1619d475b315f2cb3494747dc2424fb9856a35d6759fcf75e79995f1f1
+oid sha256:6c4152f1ae85467509d6d1bf2d3b859abf08e786a1c118f9c84f899edc2df5d1
 size 4856