Training in progress, epoch 4, checkpoint

checkpoint-324/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb94315e809cba379a98678f9eb75f0e1bf494d797f899dff6009d20abe0b2af
+oid sha256:9c50571d72cf334580391f4b758848f2e6cd915591d5d8c4d82121e57e6629bf
 size 592324828

checkpoint-324/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ea1d256984ac28cf480255343f47a06c93bbc787a0c30042dbdfc65468f92df
+oid sha256:41a9ee3412c659fa9f0538d313e6f3dd6d8aa116bd84dc1feef9426b0b92e13e
 size 1014657786

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

@@ -11,180 +11,180 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.46243291592128805,
-        "precision": 0.6588785046728972,
-        "recall": 0.3562241616903996,
-        "support": 4354.0
+        "f1-score": 0.5724454649827784,
+        "precision": 0.6002889477486154,
+        "recall": 0.5470704410796576,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8342621259029929,
-        "precision": 0.8096955128205128,
-        "recall": 0.8603661132396765,
-        "support": 2349.0
+        "f1-score": 0.7860447185813415,
+        "precision": 0.698526892771497,
+        "recall": 0.8986337593653592,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9993622448979592,
-        "precision": 0.9987253027405991,
+        "f1-score": 0.9987820830889982,
+        "precision": 0.9975671292124707,
         "recall": 1.0,
-        "support": 12536.0
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.8941052999825083,
-        "precision": 0.8401157057392676,
-        "recall": 0.955510692388216,
-        "support": 13374.0
+        "f1-score": 0.8965445389824622,
+        "precision": 0.9051258677512096,
+        "recall": 0.8881243979633962,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8857510808573268,
-      "eval_loss": 0.2754688858985901,
+      "eval_accuracy": 0.8791279948197712,
+      "eval_loss": 0.3157936632633209,
       "eval_macro avg": {
-        "f1-score": 0.7975406466761871,
-        "precision": 0.8268537564933192,
-        "recall": 0.793025241829573,
-        "support": 32613.0
-      },
-      "eval_runtime": 4.8781,
-      "eval_samples_per_second": 16.605,
-      "eval_steps_per_second": 2.255,
+        "f1-score": 0.8134542014088951,
+        "precision": 0.8003772093709481,
+        "recall": 0.8334571496021033,
+        "support": 32431.0
+      },
+      "eval_runtime": 4.8184,
+      "eval_samples_per_second": 16.603,
+      "eval_steps_per_second": 2.075,
       "eval_weighted avg": {
-        "f1-score": 0.8726239240080429,
-        "precision": 0.8746959682542021,
-        "recall": 0.8857510808573268,
-        "support": 32613.0
+        "f1-score": 0.8781740995293698,
+        "precision": 0.8793943851252293,
+        "recall": 0.8791279948197712,
+        "support": 32431.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.6446312571646924,
-        "precision": 0.5518482172064115,
-        "recall": 0.77491961414791,
-        "support": 4354.0
+        "f1-score": 0.6090644692195831,
+        "precision": 0.680108254397835,
+        "recall": 0.5514592933947773,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8912341158733577,
-        "precision": 0.9019180470793374,
-        "recall": 0.8808003405704555,
-        "support": 2349.0
+        "f1-score": 0.8585858585858586,
+        "precision": 0.8960229995208433,
+        "recall": 0.8241516086381666,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998005504806733,
-        "precision": 0.9999202106439001,
-        "recall": 0.9996809189534142,
-        "support": 12536.0
+        "f1-score": 0.999458141425088,
+        "precision": 0.9992776523702032,
+        "recall": 0.9996386956914461,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.8690409646250897,
-        "precision": 0.9324023303632625,
-        "recall": 0.8137430835950351,
-        "support": 13374.0
+        "f1-score": 0.9105885478942475,
+        "precision": 0.8801849235905997,
+        "recall": 0.9431677445988716,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8848618649004998,
-      "eval_loss": 0.25178349018096924,
+      "eval_accuracy": 0.8990780426135487,
+      "eval_loss": 0.24451124668121338,
       "eval_macro avg": {
-        "f1-score": 0.8511767220359533,
-        "precision": 0.8465222013232279,
-        "recall": 0.8672859893167036,
-        "support": 32613.0
-      },
-      "eval_runtime": 4.907,
-      "eval_samples_per_second": 16.507,
-      "eval_steps_per_second": 2.242,
+        "f1-score": 0.8444242542811943,
+        "precision": 0.8638984574698704,
+        "recall": 0.8296043355808155,
+        "support": 32431.0
+      },
+      "eval_runtime": 4.8529,
+      "eval_samples_per_second": 16.485,
+      "eval_steps_per_second": 2.061,
       "eval_weighted avg": {
-        "f1-score": 0.8909418634778542,
-        "precision": 0.9053537287957648,
-        "recall": 0.8848618649004998,
-        "support": 32613.0
+        "f1-score": 0.8949194640367888,
+        "precision": 0.8938342965082798,
+        "recall": 0.8990780426135487,
+        "support": 32431.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.6598764019226366,
-        "precision": 0.6576186131386861,
-        "recall": 0.6621497473587505,
-        "support": 4354.0
+        "f1-score": 0.6542320859402667,
+        "precision": 0.7079713847726111,
+        "recall": 0.60807548825982,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9062119366626066,
-        "precision": 0.8661233993015134,
-        "recall": 0.9501915708812261,
-        "support": 2349.0
+        "f1-score": 0.8915187376725839,
+        "precision": 0.8866608544027899,
+        "recall": 0.8964301454385192,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998404722022812,
-        "precision": 0.9997607273887382,
-        "recall": 0.9999202297383536,
-        "support": 12536.0
+        "f1-score": 0.9995936611133684,
+        "precision": 0.9992778479870013,
+        "recall": 0.9999096739228616,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9007852612503775,
-        "precision": 0.9097148086014946,
-        "recall": 0.8920293106026619,
-        "support": 13374.0
+        "f1-score": 0.9157990498332153,
+        "precision": 0.8973258501155497,
+        "recall": 0.935048850970139,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9070002759635728,
-      "eval_loss": 0.23778581619262695,
+      "eval_accuracy": 0.9085442940396534,
+      "eval_loss": 0.22665414214134216,
       "eval_macro avg": {
-        "f1-score": 0.8666785180094754,
-        "precision": 0.8583043871076081,
-        "recall": 0.876072714645248,
-        "support": 32613.0
-      },
-      "eval_runtime": 4.9168,
-      "eval_samples_per_second": 16.474,
-      "eval_steps_per_second": 2.237,
+        "f1-score": 0.8652858836398586,
+        "precision": 0.872808984319488,
+        "recall": 0.8598660396478349,
+        "support": 32431.0
+      },
+      "eval_runtime": 4.8494,
+      "eval_samples_per_second": 16.497,
+      "eval_steps_per_second": 2.062,
       "eval_weighted avg": {
-        "f1-score": 0.907089072967282,
-        "precision": 0.9075314026721463,
-        "recall": 0.9070002759635728,
-        "support": 32613.0
+        "f1-score": 0.9059516032151627,
+        "precision": 0.9047762338408377,
+        "recall": 0.9085442940396534,
+        "support": 32431.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6604639922667954,
-        "precision": 0.696838347781744,
-        "recall": 0.6276986678915939,
-        "support": 4354.0
+        "f1-score": 0.6861642294713161,
+        "precision": 0.6424741478360781,
+        "recall": 0.7362299758613122,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9140214420853479,
-        "precision": 0.9028239202657807,
-        "recall": 0.9255002128565347,
-        "support": 2349.0
+        "f1-score": 0.8995515695067264,
+        "precision": 0.9155636695572797,
+        "recall": 0.884089907448215,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9994417862838917,
-        "precision": 0.9991230867346939,
-        "recall": 0.9997606892150607,
-        "support": 12536.0
+        "f1-score": 0.9976481230212574,
+        "precision": 0.9990941208442794,
+        "recall": 0.9962063047601842,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9084203149780549,
-        "precision": 0.8963534463934785,
-        "recall": 0.9208165096455809,
-        "support": 13374.0
+        "f1-score": 0.9062532879739067,
+        "precision": 0.9242435081193219,
+        "recall": 0.8889500481629283,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.912366234323736,
-      "eval_loss": 0.2598011791706085,
+      "eval_accuracy": 0.9037649162838025,
+      "eval_loss": 0.2452327460050583,
       "eval_macro avg": {
-        "f1-score": 0.8705868839035225,
-        "precision": 0.8737847002939243,
-        "recall": 0.8684440199021926,
-        "support": 32613.0
-      },
-      "eval_runtime": 4.8872,
-      "eval_samples_per_second": 16.574,
-      "eval_steps_per_second": 2.251,
+        "f1-score": 0.8724043024933017,
+        "precision": 0.8703438615892398,
+        "recall": 0.8763690590581599,
+        "support": 32431.0
+      },
+      "eval_runtime": 4.8393,
+      "eval_samples_per_second": 16.531,
+      "eval_steps_per_second": 2.066,
       "eval_weighted avg": {
-        "f1-score": 0.9107077581074873,
-        "precision": 0.9096864919608297,
-        "recall": 0.912366234323736,
-        "support": 32613.0
+        "f1-score": 0.9060584182508294,
+        "precision": 0.9095955973231696,
+        "recall": 0.9037649162838025,
+        "support": 32431.0
       },
       "step": 324
     }
@@ -194,7 +194,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 573368735469600.0,
+  "total_flos": 575154930907200.0,
   "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 sha256:3f40e108617f6f8082cc0f1a47e89121d861cfb2e77ef6b602a909d9b6ba8d15
+oid sha256:b2dd7b387b7349843b62a9d5b4facde1d14a6a35003d55f9e21a477105ecd1d3
 size 4856