updated model weights

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/mnt/beegfs/mc000051/CERPLES/Models/roberta-base-biomedical-clinical-es",
+  "_name_or_path": "Models/roberta-base-biomedical-clinical-es",
   "architectures": [
     "RobertaForSequenceClassification"
   ],

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb0df800a0717e27a653a11d7f8346b293ebec23f5e161f45ac4940172e19941
+size 1008039837

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6e41143e7af891ffe98910ef36cc712dbb46d86b33a08904be3bd8a2543f6a
+size 504033325

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dcd8bd2facbcb38c978d6b2b10cbacd8e9f54834a513467b1196a471554682c
+oid sha256:2c92db6b019bcc5daa66264ee6196af75e462e009a9f3d24a99e49aae2e98453
 size 14567

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:052b473372c295fdcecaddd4e0fe02e5f1ad19b0e470cf4538018b0a63f4f2cb
+size 623

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

tokenizer_config.json

@@ -1 +1 @@
-{"errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": true, "trim_offsets": true, "max_len": 512, "special_tokens_map_file": null, "name_or_path": "/mnt/beegfs/mc000051/CERPLES/Models/roberta-base-biomedical-clinical-es", "tokenizer_class": "RobertaTokenizer"}
+{"errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": true, "trim_offsets": true, "max_len": 512, "special_tokens_map_file": null, "name_or_path": "Models/roberta-base-biomedical-clinical-es", "tokenizer_class": "RobertaTokenizer"}

trainer_state.json

@@ -1,156 +1,578 @@
 {
-  "best_metric": 0.7442957192288286,
-  "best_model_checkpoint": "./CARES/checkpoints/roberta-/checkpoint-3500",
-  "epoch": 49.29577464788732,
-  "global_step": 3500,
+  "best_metric": 0.7414142113821449,
+  "best_model_checkpoint": "./CARES/checkpoints/roberta/checkpoint-2698",
+  "epoch": 38.0,
+  "global_step": 2698,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.2509869933128357,
+      "eval_macro_f1": 0.04342407975460123,
+      "eval_macro_precision": 0.034867610837438424,
+      "eval_macro_recall": 0.057545731707317076,
+      "eval_micro_f1": 0.38635394456289984,
+      "eval_micro_precision": 0.5578817733990148,
+      "eval_micro_recall": 0.29549902152641877,
+      "eval_runtime": 2.7138,
+      "eval_samples_per_second": 355.964,
+      "eval_steps_per_second": 11.423,
+      "step": 71
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.21916496753692627,
+      "eval_macro_f1": 0.10341134166953839,
+      "eval_macro_precision": 0.1498316489705596,
+      "eval_macro_recall": 0.08601393938480183,
+      "eval_micro_f1": 0.42796410014171,
+      "eval_micro_precision": 0.7756849315068494,
+      "eval_micro_recall": 0.29549902152641877,
+      "eval_runtime": 2.7022,
+      "eval_samples_per_second": 357.485,
+      "eval_steps_per_second": 11.472,
+      "step": 142
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.1901940405368805,
+      "eval_macro_f1": 0.17433694331035438,
+      "eval_macro_precision": 0.1996622297635695,
+      "eval_macro_recall": 0.1554066444975929,
+      "eval_micro_f1": 0.5823627287853577,
+      "eval_micro_precision": 0.8036739380022963,
+      "eval_micro_recall": 0.45662100456621,
+      "eval_runtime": 2.7029,
+      "eval_samples_per_second": 357.398,
+      "eval_steps_per_second": 11.469,
+      "step": 213
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.16621476411819458,
+      "eval_macro_f1": 0.20771764577178875,
+      "eval_macro_precision": 0.3206539932628853,
+      "eval_macro_recall": 0.18121219739914096,
+      "eval_micro_f1": 0.6394934705184012,
+      "eval_micro_precision": 0.8128772635814889,
+      "eval_micro_recall": 0.5270711024135681,
+      "eval_runtime": 2.7016,
+      "eval_samples_per_second": 357.565,
+      "eval_steps_per_second": 11.475,
+      "step": 284
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.1592407524585724,
+      "eval_macro_f1": 0.32374988901102003,
+      "eval_macro_precision": 0.36319303902027156,
+      "eval_macro_recall": 0.30985669516000114,
+      "eval_micro_f1": 0.6886094674556215,
+      "eval_micro_precision": 0.7950469684030743,
+      "eval_micro_recall": 0.6073059360730594,
+      "eval_runtime": 2.7015,
+      "eval_samples_per_second": 357.579,
+      "eval_steps_per_second": 11.475,
+      "step": 355
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.1491735577583313,
+      "eval_macro_f1": 0.3533190399166411,
+      "eval_macro_precision": 0.4078813390257998,
+      "eval_macro_recall": 0.33542234175118296,
+      "eval_micro_f1": 0.7270094134685011,
+      "eval_micro_precision": 0.8169243287225386,
+      "eval_micro_recall": 0.6549249836921069,
+      "eval_runtime": 2.7029,
+      "eval_samples_per_second": 357.392,
+      "eval_steps_per_second": 11.469,
+      "step": 426
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.14026637375354767,
+      "eval_macro_f1": 0.47073084873229765,
+      "eval_macro_precision": 0.6837800649883113,
+      "eval_macro_recall": 0.423964742702268,
+      "eval_micro_f1": 0.75768156424581,
+      "eval_micro_precision": 0.8151765589782118,
+      "eval_micro_recall": 0.7077625570776256,
+      "eval_runtime": 2.7322,
+      "eval_samples_per_second": 353.558,
+      "eval_steps_per_second": 11.346,
+      "step": 497
+    },
     {
       "epoch": 7.04,
       "learning_rate": 4.647887323943662e-05,
-      "loss": 0.1802,
+      "loss": 0.1794,
       "step": 500
     },
     {
-      "epoch": 7.04,
-      "eval_loss": 0.1463283747434616,
-      "eval_macro_f1": 0.38060566625566594,
-      "eval_macro_precision": 0.5037224636955915,
-      "eval_macro_recall": 0.36676192196038643,
-      "eval_micro_f1": 0.7358149620527648,
-      "eval_micro_precision": 0.8249594813614263,
-      "eval_micro_recall": 0.6640574037834311,
-      "eval_runtime": 2.7054,
-      "eval_samples_per_second": 357.063,
-      "eval_steps_per_second": 11.459,
-      "step": 500
+      "epoch": 8.0,
+      "eval_loss": 0.13919810950756073,
+      "eval_macro_f1": 0.4762066395538287,
+      "eval_macro_precision": 0.6922175380127314,
+      "eval_macro_recall": 0.4248427568334097,
+      "eval_micro_f1": 0.7554770318021201,
+      "eval_micro_precision": 0.8242097147262915,
+      "eval_micro_recall": 0.6973255055446836,
+      "eval_runtime": 2.7464,
+      "eval_samples_per_second": 351.729,
+      "eval_steps_per_second": 11.287,
+      "step": 568
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.14247503876686096,
+      "eval_macro_f1": 0.49866416753622766,
+      "eval_macro_precision": 0.6531542097723891,
+      "eval_macro_recall": 0.4611636373916834,
+      "eval_micro_f1": 0.7498267498267498,
+      "eval_micro_precision": 0.7997043606799704,
+      "eval_micro_recall": 0.7058056099151989,
+      "eval_runtime": 2.7469,
+      "eval_samples_per_second": 351.665,
+      "eval_steps_per_second": 11.285,
+      "step": 639
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.14413639903068542,
+      "eval_macro_f1": 0.515652512172511,
+      "eval_macro_precision": 0.5714508837953802,
+      "eval_macro_recall": 0.494438079407737,
+      "eval_micro_f1": 0.7669675693747909,
+      "eval_micro_precision": 0.7866941015089163,
+      "eval_micro_recall": 0.7482061317677756,
+      "eval_runtime": 2.752,
+      "eval_samples_per_second": 351.019,
+      "eval_steps_per_second": 11.265,
+      "step": 710
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 0.13881583511829376,
+      "eval_macro_f1": 0.5815495436036973,
+      "eval_macro_precision": 0.6306553976911301,
+      "eval_macro_recall": 0.5606746536385887,
+      "eval_micro_f1": 0.7825223435948361,
+      "eval_micro_precision": 0.7943548387096774,
+      "eval_micro_recall": 0.7710371819960861,
+      "eval_runtime": 2.7353,
+      "eval_samples_per_second": 353.165,
+      "eval_steps_per_second": 11.333,
+      "step": 781
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 0.14889590442180634,
+      "eval_macro_f1": 0.582949261540157,
+      "eval_macro_precision": 0.6154355859721446,
+      "eval_macro_recall": 0.5660348822272816,
+      "eval_micro_f1": 0.7689724647414371,
+      "eval_micro_precision": 0.7923875432525952,
+      "eval_micro_recall": 0.7469015003261579,
+      "eval_runtime": 2.7291,
+      "eval_samples_per_second": 353.967,
+      "eval_steps_per_second": 11.359,
+      "step": 852
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 0.14177829027175903,
+      "eval_macro_f1": 0.6003936149639709,
+      "eval_macro_precision": 0.6178459654131079,
+      "eval_macro_recall": 0.5986040907728625,
+      "eval_micro_f1": 0.7824089268132589,
+      "eval_micro_precision": 0.7873183619550859,
+      "eval_micro_recall": 0.7775603392041748,
+      "eval_runtime": 2.7236,
+      "eval_samples_per_second": 354.674,
+      "eval_steps_per_second": 11.382,
+      "step": 923
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 0.13778340816497803,
+      "eval_macro_f1": 0.6071892145561097,
+      "eval_macro_precision": 0.6420473960894854,
+      "eval_macro_recall": 0.5902771317099681,
+      "eval_micro_f1": 0.7962716378162449,
+      "eval_micro_precision": 0.8130523453433038,
+      "eval_micro_recall": 0.7801696020874103,
+      "eval_runtime": 4.6013,
+      "eval_samples_per_second": 209.942,
+      "eval_steps_per_second": 6.737,
+      "step": 994
     },
     {
       "epoch": 14.08,
       "learning_rate": 4.295774647887324e-05,
-      "loss": 0.0484,
+      "loss": 0.0459,
       "step": 1000
     },
     {
-      "epoch": 14.08,
-      "eval_loss": 0.14033463597297668,
-      "eval_macro_f1": 0.6067981552982502,
-      "eval_macro_precision": 0.6486392263482452,
-      "eval_macro_recall": 0.5884363376224964,
-      "eval_micro_f1": 0.7908386662175816,
-      "eval_micro_precision": 0.8175487465181058,
-      "eval_micro_recall": 0.7658186562296151,
-      "eval_runtime": 2.6975,
-      "eval_samples_per_second": 358.108,
-      "eval_steps_per_second": 11.492,
-      "step": 1000
+      "epoch": 15.0,
+      "eval_loss": 0.14022594690322876,
+      "eval_macro_f1": 0.6215256196989195,
+      "eval_macro_precision": 0.7025580891227308,
+      "eval_macro_recall": 0.5957635875240769,
+      "eval_micro_f1": 0.7986776859504131,
+      "eval_micro_precision": 0.8096514745308311,
+      "eval_micro_recall": 0.7879973907371167,
+      "eval_runtime": 2.7033,
+      "eval_samples_per_second": 357.348,
+      "eval_steps_per_second": 11.468,
+      "step": 1065
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 0.1448926478624344,
+      "eval_macro_f1": 0.608747774411539,
+      "eval_macro_precision": 0.621954970496014,
+      "eval_macro_recall": 0.6062377476848362,
+      "eval_micro_f1": 0.79816813869807,
+      "eval_micro_precision": 0.800524934383202,
+      "eval_micro_recall": 0.7958251793868232,
+      "eval_runtime": 2.7015,
+      "eval_samples_per_second": 357.574,
+      "eval_steps_per_second": 11.475,
+      "step": 1136
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 0.1469811201095581,
+      "eval_macro_f1": 0.6153866789938162,
+      "eval_macro_precision": 0.650591338815516,
+      "eval_macro_recall": 0.5950841704161889,
+      "eval_micro_f1": 0.7996005326231691,
+      "eval_micro_precision": 0.8164513936097892,
+      "eval_micro_recall": 0.7834311806914547,
+      "eval_runtime": 2.701,
+      "eval_samples_per_second": 357.647,
+      "eval_steps_per_second": 11.477,
+      "step": 1207
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 0.15327712893486023,
+      "eval_macro_f1": 0.6316443999429009,
+      "eval_macro_precision": 0.6512229785159827,
+      "eval_macro_recall": 0.6349187975925275,
+      "eval_micro_f1": 0.7879777850375694,
+      "eval_micro_precision": 0.7892670157068062,
+      "eval_micro_recall": 0.786692759295499,
+      "eval_runtime": 5.7579,
+      "eval_samples_per_second": 167.769,
+      "eval_steps_per_second": 5.384,
+      "step": 1278
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 0.15278153121471405,
+      "eval_macro_f1": 0.6310003042032536,
+      "eval_macro_precision": 0.6839818740829384,
+      "eval_macro_recall": 0.6255359551717035,
+      "eval_micro_f1": 0.7923659098387628,
+      "eval_micro_precision": 0.799468791500664,
+      "eval_micro_recall": 0.7853881278538812,
+      "eval_runtime": 2.7,
+      "eval_samples_per_second": 357.774,
+      "eval_steps_per_second": 11.481,
+      "step": 1349
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 0.15630246698856354,
+      "eval_macro_f1": 0.632349332888396,
+      "eval_macro_precision": 0.6992914076324832,
+      "eval_macro_recall": 0.6115304182846466,
+      "eval_micro_f1": 0.7856437273625287,
+      "eval_micro_precision": 0.793218085106383,
+      "eval_micro_recall": 0.7782126549249837,
+      "eval_runtime": 2.7011,
+      "eval_samples_per_second": 357.638,
+      "eval_steps_per_second": 11.477,
+      "step": 1420
+    },
+    {
+      "epoch": 21.0,
+      "eval_loss": 0.15509752929210663,
+      "eval_macro_f1": 0.6092726479340159,
+      "eval_macro_precision": 0.6680318673287149,
+      "eval_macro_recall": 0.5717336891200013,
+      "eval_micro_f1": 0.7991967871485944,
+      "eval_micro_precision": 0.8206185567010309,
+      "eval_micro_recall": 0.7788649706457925,
+      "eval_runtime": 2.6986,
+      "eval_samples_per_second": 357.967,
+      "eval_steps_per_second": 11.488,
+      "step": 1491
     },
     {
       "epoch": 21.13,
       "learning_rate": 3.943661971830986e-05,
-      "loss": 0.0179,
+      "loss": 0.0169,
       "step": 1500
     },
     {
-      "epoch": 21.13,
-      "eval_loss": 0.16313208639621735,
-      "eval_macro_f1": 0.6876414143457183,
-      "eval_macro_precision": 0.7316617517332842,
-      "eval_macro_recall": 0.7019290034007357,
-      "eval_micro_f1": 0.7892204042348412,
-      "eval_micro_precision": 0.7765151515151515,
+      "epoch": 22.0,
+      "eval_loss": 0.15633134543895721,
+      "eval_macro_f1": 0.6559542184820522,
+      "eval_macro_precision": 0.6920252568569637,
+      "eval_macro_recall": 0.6423031165953985,
+      "eval_micro_f1": 0.8009213557091149,
+      "eval_micro_precision": 0.8081009296148738,
+      "eval_micro_recall": 0.7938682322243966,
+      "eval_runtime": 2.7117,
+      "eval_samples_per_second": 356.232,
+      "eval_steps_per_second": 11.432,
+      "step": 1562
+    },
+    {
+      "epoch": 23.0,
+      "eval_loss": 0.16708678007125854,
+      "eval_macro_f1": 0.6341964700327825,
+      "eval_macro_precision": 0.6730301960786685,
+      "eval_macro_recall": 0.6361571334823197,
+      "eval_micro_f1": 0.7845195145949493,
+      "eval_micro_precision": 0.7889182058047494,
+      "eval_micro_recall": 0.7801696020874103,
+      "eval_runtime": 2.7072,
+      "eval_samples_per_second": 356.825,
+      "eval_steps_per_second": 11.451,
+      "step": 1633
+    },
+    {
+      "epoch": 24.0,
+      "eval_loss": 0.15924513339996338,
+      "eval_macro_f1": 0.6567583708222104,
+      "eval_macro_precision": 0.6932259526572662,
+      "eval_macro_recall": 0.6481414988629403,
+      "eval_micro_f1": 0.8018252933507171,
+      "eval_micro_precision": 0.8013029315960912,
       "eval_micro_recall": 0.8023483365949119,
-      "eval_runtime": 2.6931,
-      "eval_samples_per_second": 358.699,
-      "eval_steps_per_second": 11.511,
-      "step": 1500
+      "eval_runtime": 2.7129,
+      "eval_samples_per_second": 356.081,
+      "eval_steps_per_second": 11.427,
+      "step": 1704
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 0.15955589711666107,
+      "eval_macro_f1": 0.6499053948607745,
+      "eval_macro_precision": 0.7063535942579273,
+      "eval_macro_recall": 0.622573757353082,
+      "eval_micro_f1": 0.8029100529100529,
+      "eval_micro_precision": 0.8142186452045607,
+      "eval_micro_recall": 0.79191128506197,
+      "eval_runtime": 2.7378,
+      "eval_samples_per_second": 352.84,
+      "eval_steps_per_second": 11.323,
+      "step": 1775
+    },
+    {
+      "epoch": 26.0,
+      "eval_loss": 0.1635105311870575,
+      "eval_macro_f1": 0.6483066019452002,
+      "eval_macro_precision": 0.6824077806874724,
+      "eval_macro_recall": 0.6413983597695092,
+      "eval_micro_f1": 0.7965879265091863,
+      "eval_micro_precision": 0.8013201320132013,
+      "eval_micro_recall": 0.79191128506197,
+      "eval_runtime": 2.7006,
+      "eval_samples_per_second": 357.694,
+      "eval_steps_per_second": 11.479,
+      "step": 1846
+    },
+    {
+      "epoch": 27.0,
+      "eval_loss": 0.16744764149188995,
+      "eval_macro_f1": 0.6758901394131414,
+      "eval_macro_precision": 0.7249401901181491,
+      "eval_macro_recall": 0.6772836080088236,
+      "eval_micro_f1": 0.789776357827476,
+      "eval_micro_precision": 0.7739511584220413,
+      "eval_micro_recall": 0.8062622309197651,
+      "eval_runtime": 2.7031,
+      "eval_samples_per_second": 357.373,
+      "eval_steps_per_second": 11.468,
+      "step": 1917
+    },
+    {
+      "epoch": 28.0,
+      "eval_loss": 0.17063026130199432,
+      "eval_macro_f1": 0.6882911714431834,
+      "eval_macro_precision": 0.7320265820262954,
+      "eval_macro_recall": 0.6788510173157892,
+      "eval_micro_f1": 0.7945659377070908,
+      "eval_micro_precision": 0.8074074074074075,
+      "eval_micro_recall": 0.7821265492498369,
+      "eval_runtime": 2.7065,
+      "eval_samples_per_second": 356.912,
+      "eval_steps_per_second": 11.454,
+      "step": 1988
     },
     {
       "epoch": 28.17,
       "learning_rate": 3.5915492957746486e-05,
-      "loss": 0.0095,
+      "loss": 0.0088,
       "step": 2000
     },
     {
-      "epoch": 28.17,
-      "eval_loss": 0.1628233790397644,
-      "eval_macro_f1": 0.7285510286916194,
-      "eval_macro_precision": 0.8247103134528249,
-      "eval_macro_recall": 0.6865275313487541,
-      "eval_micro_f1": 0.8138990978950885,
-      "eval_micro_precision": 0.8342465753424657,
-      "eval_micro_recall": 0.7945205479452054,
-      "eval_runtime": 2.6912,
-      "eval_samples_per_second": 358.946,
-      "eval_steps_per_second": 11.519,
-      "step": 2000
+      "epoch": 29.0,
+      "eval_loss": 0.17283257842063904,
+      "eval_macro_f1": 0.6771985368510574,
+      "eval_macro_precision": 0.7332109505236407,
+      "eval_macro_recall": 0.663425919860526,
+      "eval_micro_f1": 0.7905294311081881,
+      "eval_micro_precision": 0.7970822281167109,
+      "eval_micro_recall": 0.7840834964122635,
+      "eval_runtime": 2.7052,
+      "eval_samples_per_second": 357.094,
+      "eval_steps_per_second": 11.46,
+      "step": 2059
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 0.18444736301898956,
+      "eval_macro_f1": 0.6746367426025414,
+      "eval_macro_precision": 0.7201523247516791,
+      "eval_macro_recall": 0.6728013388910854,
+      "eval_micro_f1": 0.7788242936018187,
+      "eval_micro_precision": 0.7755498059508409,
+      "eval_micro_recall": 0.7821265492498369,
+      "eval_runtime": 2.704,
+      "eval_samples_per_second": 357.254,
+      "eval_steps_per_second": 11.465,
+      "step": 2130
+    },
+    {
+      "epoch": 31.0,
+      "eval_loss": 0.1695714294910431,
+      "eval_macro_f1": 0.6879346286499437,
+      "eval_macro_precision": 0.7516673286655658,
+      "eval_macro_recall": 0.6669919763081706,
+      "eval_micro_f1": 0.802747791952895,
+      "eval_micro_precision": 0.8051181102362205,
+      "eval_micro_recall": 0.8003913894324853,
+      "eval_runtime": 2.7049,
+      "eval_samples_per_second": 357.124,
+      "eval_steps_per_second": 11.46,
+      "step": 2201
+    },
+    {
+      "epoch": 32.0,
+      "eval_loss": 0.16994765400886536,
+      "eval_macro_f1": 0.7053100252741045,
+      "eval_macro_precision": 0.805278845304122,
+      "eval_macro_recall": 0.6824655471979887,
+      "eval_micro_f1": 0.8006472491909384,
+      "eval_micro_precision": 0.7944765574823378,
+      "eval_micro_recall": 0.806914546640574,
+      "eval_runtime": 2.7034,
+      "eval_samples_per_second": 357.325,
+      "eval_steps_per_second": 11.467,
+      "step": 2272
+    },
+    {
+      "epoch": 33.0,
+      "eval_loss": 0.17551660537719727,
+      "eval_macro_f1": 0.7235647899419184,
+      "eval_macro_precision": 0.7984745552617538,
+      "eval_macro_recall": 0.6960462941676209,
+      "eval_micro_f1": 0.8010352636687157,
+      "eval_micro_precision": 0.7946084724005135,
+      "eval_micro_recall": 0.8075668623613829,
+      "eval_runtime": 2.7047,
+      "eval_samples_per_second": 357.151,
+      "eval_steps_per_second": 11.461,
+      "step": 2343
+    },
+    {
+      "epoch": 34.0,
+      "eval_loss": 0.17244330048561096,
+      "eval_macro_f1": 0.6948832290211537,
+      "eval_macro_precision": 0.7553961335643664,
+      "eval_macro_recall": 0.6730022583492952,
+      "eval_micro_f1": 0.8063989552726085,
+      "eval_micro_precision": 0.8071895424836601,
+      "eval_micro_recall": 0.8056099151989563,
+      "eval_runtime": 2.7035,
+      "eval_samples_per_second": 357.318,
+      "eval_steps_per_second": 11.467,
+      "step": 2414
+    },
+    {
+      "epoch": 35.0,
+      "eval_loss": 0.17627869546413422,
+      "eval_macro_f1": 0.7275662557839568,
+      "eval_macro_precision": 0.7750438946995515,
+      "eval_macro_recall": 0.717891607952716,
+      "eval_micro_f1": 0.8019261637239165,
+      "eval_micro_precision": 0.7895069532237674,
+      "eval_micro_recall": 0.8147423352902805,
+      "eval_runtime": 2.7043,
+      "eval_samples_per_second": 357.211,
+      "eval_steps_per_second": 11.463,
+      "step": 2485
     },
     {
       "epoch": 35.21,
       "learning_rate": 3.23943661971831e-05,
-      "loss": 0.0055,
+      "loss": 0.0051,
       "step": 2500
     },
     {
-      "epoch": 35.21,
-      "eval_loss": 0.1734953373670578,
-      "eval_macro_f1": 0.7421994453880427,
-      "eval_macro_precision": 0.8204432806900727,
-      "eval_macro_recall": 0.7088284387134485,
-      "eval_micro_f1": 0.8176722716782063,
-      "eval_micro_precision": 0.8266666666666667,
-      "eval_micro_recall": 0.8088714938030006,
-      "eval_runtime": 2.6929,
-      "eval_samples_per_second": 358.718,
-      "eval_steps_per_second": 11.512,
-      "step": 2500
+      "epoch": 36.0,
+      "eval_loss": 0.1767842173576355,
+      "eval_macro_f1": 0.6877210668749649,
+      "eval_macro_precision": 0.7524115221848844,
+      "eval_macro_recall": 0.6652074916473143,
+      "eval_micro_f1": 0.7996077149395227,
+      "eval_micro_precision": 0.8014416775884666,
+      "eval_micro_recall": 0.7977821265492498,
+      "eval_runtime": 2.7004,
+      "eval_samples_per_second": 357.724,
+      "eval_steps_per_second": 11.48,
+      "step": 2556
+    },
+    {
+      "epoch": 37.0,
+      "eval_loss": 0.17860282957553864,
+      "eval_macro_f1": 0.7341275703601238,
+      "eval_macro_precision": 0.8208336935763849,
+      "eval_macro_recall": 0.6963532745066767,
+      "eval_micro_f1": 0.8019512195121952,
+      "eval_micro_precision": 0.7996108949416343,
+      "eval_micro_recall": 0.8043052837573386,
+      "eval_runtime": 2.7018,
+      "eval_samples_per_second": 357.539,
+      "eval_steps_per_second": 11.474,
+      "step": 2627
     },
     {
-      "epoch": 42.25,
-      "learning_rate": 2.887323943661972e-05,
-      "loss": 0.0039,
-      "step": 3000
-    },
-    {
-      "epoch": 42.25,
-      "eval_loss": 0.1867484599351883,
-      "eval_macro_f1": 0.7349168384365183,
-      "eval_macro_precision": 0.7741277072189265,
-      "eval_macro_recall": 0.7306178466824944,
-      "eval_micro_f1": 0.8012924071082391,
-      "eval_micro_precision": 0.793854033290653,
-      "eval_micro_recall": 0.8088714938030006,
-      "eval_runtime": 2.6968,
-      "eval_samples_per_second": 358.208,
-      "eval_steps_per_second": 11.495,
-      "step": 3000
-    },
-    {
-      "epoch": 49.3,
-      "learning_rate": 2.535211267605634e-05,
-      "loss": 0.0032,
-      "step": 3500
-    },
-    {
-      "epoch": 49.3,
-      "eval_loss": 0.181275874376297,
-      "eval_macro_f1": 0.7442957192288286,
-      "eval_macro_precision": 0.7848363899123905,
-      "eval_macro_recall": 0.7360907675141599,
-      "eval_micro_f1": 0.8202794930126746,
-      "eval_micro_precision": 0.8173575129533679,
-      "eval_micro_recall": 0.8232224396607958,
-      "eval_runtime": 2.6934,
-      "eval_samples_per_second": 358.652,
-      "eval_steps_per_second": 11.51,
-      "step": 3500
+      "epoch": 38.0,
+      "eval_loss": 0.18057870864868164,
+      "eval_macro_f1": 0.7414142113821449,
+      "eval_macro_precision": 0.8229012131281167,
+      "eval_macro_recall": 0.7070558641306933,
+      "eval_micro_f1": 0.8049575994781474,
+      "eval_micro_precision": 0.8049575994781474,
+      "eval_micro_recall": 0.8049575994781474,
+      "eval_runtime": 2.703,
+      "eval_samples_per_second": 357.385,
+      "eval_steps_per_second": 11.469,
+      "step": 2698
     }
   ],
   "max_steps": 7100,
   "num_train_epochs": 100,
-  "total_flos": 2.877157614419203e+16,
+  "total_flos": 2.2147253694876576e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95ef9e04b1c54450ebb0e5d368ad5f45d662a1cb072d8e0e194111a56c16c7d5
-size 3247
+oid sha256:3415febb816dc8caa278fa1560a2ebbb18bac1da83bdbbd581a6a1409f9c4b4d
+size 3183