Update model

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "toxicity_c_202201171730/checkpoint-2500",
+  "_name_or_path": "toxicity_c_202201181030/checkpoint-3300",
   "architectures": [
     "RobertaForSequenceClassification"
   ],

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:910daa14f01d5ccf57987683c512c11dfd9ebed44e78c4bfcd01b03c71999102
+oid sha256:98d4743bd03d3c1ad340a17658f87635a81f2de0e0ebc2159783b68aa8ea22b3
 size 934225693

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e6a677f77a0228f884d171751089e81f9869289f3b9e606ba144d88cd0a92bf
+oid sha256:0355d31ebf383df36d476633e5e4cd99f9da175fc28da8badece785b58ed5c07
 size 467137197

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c71aa49f7fbc9ee34a4ce29c3f8b22493dd0074f4e95650db7ac5ff4b84de20
+oid sha256:17362cbb3e28fd8e671ec4531f2848331bee2a708eb2449b45aee7f5cc68841f
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57471df008f47fc442d96a15e0831d3c5089476f7e646f62a2d9599992ca66f3
+oid sha256:16e017439bb342f037136f336301ac4b2f603da5f60ef2b9639ffe4c9bb1be2f
 size 623

tokenizer_config.json

@@ -1 +1 @@
-{"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "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"}, "add_prefix_space": false, "errors": "replace", "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"}, "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"}, "trim_offsets": true, "special_tokens_map_file": "./robbert-v2-dutch-base/special_tokens_map.json", "name_or_path": "toxicity_c_202201171730/checkpoint-2500", "tokenizer_class": "RobertaTokenizer"}
+{"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "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"}, "add_prefix_space": false, "errors": "replace", "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"}, "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"}, "trim_offsets": true, "special_tokens_map_file": "./robbert-v2-dutch-base/special_tokens_map.json", "name_or_path": "toxicity_c_202201181030/checkpoint-3300", "tokenizer_class": "RobertaTokenizer"}

trainer_state.json

@@ -1,256 +1,610 @@
 {
-  "best_metric": 0.5771182141045155,
-  "best_model_checkpoint": "./results_c/checkpoint-2250",
-  "epoch": 0.7100031555695803,
-  "global_step": 2250,
+  "best_metric": 0.8023715415019763,
+  "best_model_checkpoint": "./toxicity_c_202201181030/checkpoint-3000",
+  "epoch": 1.2414895617829602,
+  "global_step": 3300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.03,
-      "learning_rate": 9.600000000000001e-06,
-      "loss": 0.2758,
+      "epoch": 0.04,
+      "learning_rate": 2.94e-05,
+      "loss": 0.2336,
       "step": 100
     },
     {
-      "epoch": 0.06,
-      "learning_rate": 9.200000000000002e-06,
-      "loss": 0.2004,
+      "epoch": 0.04,
+      "eval_accuracy": 0.9445713034502174,
+      "eval_f1": 0.6696495152870993,
+      "eval_loss": 0.1531878262758255,
+      "eval_precision": 0.8655421686746988,
+      "eval_recall": 0.5460626330191548,
+      "eval_runtime": 209.3715,
+      "eval_samples_per_second": 152.69,
+      "eval_steps_per_second": 19.09,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.88e-05,
+      "loss": 0.1549,
       "step": 200
     },
     {
       "epoch": 0.08,
-      "eval_accuracy": 0.9320512236248462,
-      "eval_f1": 0.5911708253358925,
-      "eval_loss": 0.17533589899539948,
-      "eval_precision": 0.6431980906921241,
-      "eval_recall": 0.5469304921359716,
-      "eval_runtime": 153.0891,
-      "eval_samples_per_second": 286.67,
-      "eval_steps_per_second": 35.835,
-      "step": 250
-    },
-    {
-      "epoch": 0.09,
-      "learning_rate": 8.8e-06,
-      "loss": 0.1794,
+      "eval_accuracy": 0.9456973943507773,
+      "eval_f1": 0.6756352765321375,
+      "eval_loss": 0.1629406213760376,
+      "eval_precision": 0.8763936015511391,
+      "eval_recall": 0.5497111584068106,
+      "eval_runtime": 209.809,
+      "eval_samples_per_second": 152.372,
+      "eval_steps_per_second": 19.051,
+      "step": 200
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.8199999999999998e-05,
+      "loss": 0.158,
       "step": 300
     },
     {
-      "epoch": 0.13,
-      "learning_rate": 8.400000000000001e-06,
-      "loss": 0.1954,
+      "epoch": 0.11,
+      "eval_accuracy": 0.949920235227877,
+      "eval_f1": 0.7275820997107366,
+      "eval_loss": 0.13407668471336365,
+      "eval_precision": 0.8261205564142194,
+      "eval_recall": 0.6500456065673457,
+      "eval_runtime": 210.4551,
+      "eval_samples_per_second": 151.904,
+      "eval_steps_per_second": 18.992,
+      "step": 300
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 2.7600000000000003e-05,
+      "loss": 0.1321,
       "step": 400
     },
     {
-      "epoch": 0.16,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 0.1774,
-      "step": 500
+      "epoch": 0.15,
+      "eval_accuracy": 0.9515468109731302,
+      "eval_f1": 0.7355301348813386,
+      "eval_loss": 0.1435689479112625,
+      "eval_precision": 0.8387850467289719,
+      "eval_recall": 0.6549103070842202,
+      "eval_runtime": 209.6366,
+      "eval_samples_per_second": 152.497,
+      "eval_steps_per_second": 19.066,
+      "step": 400
     },
     {
-      "epoch": 0.16,
-      "eval_accuracy": 0.9371097844415075,
-      "eval_f1": 0.5845875978326308,
-      "eval_loss": 0.17303350567817688,
-      "eval_precision": 0.7187268689859363,
-      "eval_recall": 0.4926433282597666,
-      "eval_runtime": 153.4653,
-      "eval_samples_per_second": 285.967,
-      "eval_steps_per_second": 35.747,
+      "epoch": 0.19,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.1408,
       "step": 500
     },
     {
       "epoch": 0.19,
-      "learning_rate": 7.600000000000001e-06,
-      "loss": 0.1597,
+      "eval_accuracy": 0.9473865307016172,
+      "eval_f1": 0.7431276725717777,
+      "eval_loss": 0.13682132959365845,
+      "eval_precision": 0.7465480208652961,
+      "eval_recall": 0.739738522347218,
+      "eval_runtime": 210.8965,
+      "eval_samples_per_second": 151.586,
+      "eval_steps_per_second": 18.952,
+      "step": 500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.64e-05,
+      "loss": 0.1385,
       "step": 600
     },
     {
-      "epoch": 0.22,
-      "learning_rate": 7.2000000000000005e-06,
-      "loss": 0.1741,
+      "epoch": 0.23,
+      "eval_accuracy": 0.9507960837060903,
+      "eval_f1": 0.7527118377613583,
+      "eval_loss": 0.12854768335819244,
+      "eval_precision": 0.779296875,
+      "eval_recall": 0.7278808148373366,
+      "eval_runtime": 210.2245,
+      "eval_samples_per_second": 152.071,
+      "eval_steps_per_second": 19.013,
+      "step": 600
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 2.58e-05,
+      "loss": 0.1313,
       "step": 700
     },
     {
-      "epoch": 0.24,
-      "eval_accuracy": 0.9397757827097479,
-      "eval_f1": 0.5786704925872788,
-      "eval_loss": 0.16740337014198303,
-      "eval_precision": 0.7786357786357786,
-      "eval_recall": 0.4604261796042618,
-      "eval_runtime": 152.7876,
-      "eval_samples_per_second": 287.235,
-      "eval_steps_per_second": 35.906,
-      "step": 750
+      "epoch": 0.26,
+      "eval_accuracy": 0.9529857049016235,
+      "eval_f1": 0.7370078740157481,
+      "eval_loss": 0.14518482983112335,
+      "eval_precision": 0.8680956306677658,
+      "eval_recall": 0.6403162055335968,
+      "eval_runtime": 210.6173,
+      "eval_samples_per_second": 151.787,
+      "eval_steps_per_second": 18.978,
+      "step": 700
     },
     {
-      "epoch": 0.25,
-      "learning_rate": 6.800000000000001e-06,
-      "loss": 0.1676,
+      "epoch": 0.3,
+      "learning_rate": 2.52e-05,
+      "loss": 0.1399,
       "step": 800
     },
     {
-      "epoch": 0.28,
-      "learning_rate": 6.4000000000000006e-06,
-      "loss": 0.1694,
+      "epoch": 0.3,
+      "eval_accuracy": 0.949826394319497,
+      "eval_f1": 0.7610250297973777,
+      "eval_loss": 0.13412030041217804,
+      "eval_precision": 0.7461291264972246,
+      "eval_recall": 0.7765278200060809,
+      "eval_runtime": 209.9319,
+      "eval_samples_per_second": 152.283,
+      "eval_steps_per_second": 19.04,
+      "step": 800
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.4599999999999998e-05,
+      "loss": 0.1272,
+      "step": 900
+    },
+    {
+      "epoch": 0.34,
+      "eval_accuracy": 0.9536425912602834,
+      "eval_f1": 0.7640878701050622,
+      "eval_loss": 0.12946395576000214,
+      "eval_precision": 0.8018710324089542,
+      "eval_recall": 0.7297050775311645,
+      "eval_runtime": 210.1105,
+      "eval_samples_per_second": 152.153,
+      "eval_steps_per_second": 19.023,
       "step": 900
     },
     {
-      "epoch": 0.32,
-      "learning_rate": 6e-06,
-      "loss": 0.1605,
+      "epoch": 0.38,
+      "learning_rate": 2.4e-05,
+      "loss": 0.1324,
       "step": 1000
     },
     {
-      "epoch": 0.32,
-      "eval_accuracy": 0.9402770815294171,
-      "eval_f1": 0.5811091577433275,
-      "eval_loss": 0.16347511112689972,
-      "eval_precision": 0.7853131749460043,
-      "eval_recall": 0.4611872146118721,
-      "eval_runtime": 153.4862,
-      "eval_samples_per_second": 285.928,
-      "eval_steps_per_second": 35.743,
+      "epoch": 0.38,
+      "eval_accuracy": 0.9478244549407239,
+      "eval_f1": 0.7562116340251388,
+      "eval_loss": 0.14241644740104675,
+      "eval_precision": 0.7281170841542358,
+      "eval_recall": 0.7865612648221344,
+      "eval_runtime": 210.5453,
+      "eval_samples_per_second": 151.839,
+      "eval_steps_per_second": 18.984,
       "step": 1000
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 5.600000000000001e-06,
-      "loss": 0.1659,
+      "epoch": 0.41,
+      "learning_rate": 2.3400000000000003e-05,
+      "loss": 0.1198,
       "step": 1100
     },
     {
-      "epoch": 0.38,
-      "learning_rate": 5.2e-06,
-      "loss": 0.1524,
+      "epoch": 0.41,
+      "eval_accuracy": 0.9530482655072101,
+      "eval_f1": 0.7681137030743087,
+      "eval_loss": 0.1265845000743866,
+      "eval_precision": 0.7807788944723618,
+      "eval_recall": 0.7558528428093646,
+      "eval_runtime": 210.173,
+      "eval_samples_per_second": 152.108,
+      "eval_steps_per_second": 19.018,
+      "step": 1100
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.2800000000000002e-05,
+      "loss": 0.1232,
       "step": 1200
     },
     {
-      "epoch": 0.39,
-      "eval_accuracy": 0.9379300916009662,
-      "eval_f1": 0.5199154035953473,
-      "eval_loss": 0.1782103329896927,
-      "eval_precision": 0.8516166281755196,
-      "eval_recall": 0.3741755454084221,
-      "eval_runtime": 153.3211,
-      "eval_samples_per_second": 286.236,
-      "eval_steps_per_second": 35.781,
-      "step": 1250
+      "epoch": 0.45,
+      "eval_accuracy": 0.9558634927586099,
+      "eval_f1": 0.7611308616895208,
+      "eval_loss": 0.11988817900419235,
+      "eval_precision": 0.8586707410236822,
+      "eval_recall": 0.6834904226208574,
+      "eval_runtime": 209.5961,
+      "eval_samples_per_second": 152.527,
+      "eval_steps_per_second": 19.07,
+      "step": 1200
     },
     {
-      "epoch": 0.41,
-      "learning_rate": 4.800000000000001e-06,
-      "loss": 0.1564,
+      "epoch": 0.49,
+      "learning_rate": 2.22e-05,
+      "loss": 0.1239,
+      "step": 1300
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.9564578185116832,
+      "eval_f1": 0.7667560321715818,
+      "eval_loss": 0.12424681335687637,
+      "eval_precision": 0.8540500186636805,
+      "eval_recall": 0.6956521739130435,
+      "eval_runtime": 210.1306,
+      "eval_samples_per_second": 152.139,
+      "eval_steps_per_second": 19.022,
       "step": 1300
     },
     {
-      "epoch": 0.44,
-      "learning_rate": 4.4e-06,
-      "loss": 0.1685,
+      "epoch": 0.53,
+      "learning_rate": 2.16e-05,
+      "loss": 0.1236,
+      "step": 1400
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.9570834245675498,
+      "eval_f1": 0.7704918032786885,
+      "eval_loss": 0.1350642293691635,
+      "eval_precision": 0.8564522127184827,
+      "eval_recall": 0.7002128306476133,
+      "eval_runtime": 210.8014,
+      "eval_samples_per_second": 151.655,
+      "eval_steps_per_second": 18.961,
       "step": 1400
     },
     {
-      "epoch": 0.47,
-      "learning_rate": 4.000000000000001e-06,
-      "loss": 0.1536,
+      "epoch": 0.56,
+      "learning_rate": 2.1e-05,
+      "loss": 0.1254,
       "step": 1500
     },
     {
-      "epoch": 0.47,
-      "eval_accuracy": 0.9428291482477328,
-      "eval_f1": 0.642643498077197,
-      "eval_loss": 0.14973676204681396,
-      "eval_precision": 0.7327054238389087,
-      "eval_recall": 0.5722983257229832,
-      "eval_runtime": 153.7107,
-      "eval_samples_per_second": 285.51,
-      "eval_steps_per_second": 35.69,
+      "epoch": 0.56,
+      "eval_accuracy": 0.9575526291094498,
+      "eval_f1": 0.7775774463202754,
+      "eval_loss": 0.11828587204217911,
+      "eval_precision": 0.8435277382645804,
+      "eval_recall": 0.7211918516266342,
+      "eval_runtime": 209.8683,
+      "eval_samples_per_second": 152.329,
+      "eval_steps_per_second": 19.045,
       "step": 1500
     },
     {
-      "epoch": 0.5,
-      "learning_rate": 3.6000000000000003e-06,
-      "loss": 0.1698,
+      "epoch": 0.6,
+      "learning_rate": 2.04e-05,
+      "loss": 0.1252,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6,
+      "eval_accuracy": 0.9568644624479965,
+      "eval_f1": 0.7638294228463779,
+      "eval_loss": 0.12166079878807068,
+      "eval_precision": 0.8745098039215686,
+      "eval_recall": 0.6780176345393737,
+      "eval_runtime": 210.147,
+      "eval_samples_per_second": 152.127,
+      "eval_steps_per_second": 19.02,
       "step": 1600
     },
     {
-      "epoch": 0.54,
-      "learning_rate": 3.2000000000000003e-06,
-      "loss": 0.1534,
+      "epoch": 0.64,
+      "learning_rate": 1.98e-05,
+      "loss": 0.1256,
       "step": 1700
     },
     {
-      "epoch": 0.55,
-      "eval_accuracy": 0.9436038827872214,
-      "eval_f1": 0.6217331499312242,
-      "eval_loss": 0.15507051348686218,
-      "eval_precision": 0.7820069204152249,
-      "eval_recall": 0.5159817351598174,
-      "eval_runtime": 153.2946,
-      "eval_samples_per_second": 286.285,
-      "eval_steps_per_second": 35.787,
-      "step": 1750
+      "epoch": 0.64,
+      "eval_accuracy": 0.9574275078982765,
+      "eval_f1": 0.7709910819451455,
+      "eval_loss": 0.11569029092788696,
+      "eval_precision": 0.8632253202712886,
+      "eval_recall": 0.6965643052599574,
+      "eval_runtime": 210.5963,
+      "eval_samples_per_second": 151.802,
+      "eval_steps_per_second": 18.979,
+      "step": 1700
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 2.8000000000000003e-06,
-      "loss": 0.1637,
+      "epoch": 0.68,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.1324,
       "step": 1800
     },
     {
-      "epoch": 0.6,
-      "learning_rate": 2.4000000000000003e-06,
-      "loss": 0.1575,
+      "epoch": 0.68,
+      "eval_accuracy": 0.95426819731615,
+      "eval_f1": 0.7747303543913714,
+      "eval_loss": 0.12124760448932648,
+      "eval_precision": 0.7853795688847235,
+      "eval_recall": 0.7643660687138948,
+      "eval_runtime": 210.0339,
+      "eval_samples_per_second": 152.209,
+      "eval_steps_per_second": 19.03,
+      "step": 1800
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.86e-05,
+      "loss": 0.1192,
       "step": 1900
     },
     {
-      "epoch": 0.63,
-      "learning_rate": 2.0000000000000003e-06,
-      "loss": 0.1669,
+      "epoch": 0.71,
+      "eval_accuracy": 0.9576777503206231,
+      "eval_f1": 0.7719534805326143,
+      "eval_loss": 0.12099746614694595,
+      "eval_precision": 0.8661119515885023,
+      "eval_recall": 0.6962602614776527,
+      "eval_runtime": 209.8273,
+      "eval_samples_per_second": 152.359,
+      "eval_steps_per_second": 19.049,
+      "step": 1900
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8e-05,
+      "loss": 0.1234,
       "step": 2000
     },
     {
-      "epoch": 0.63,
-      "eval_accuracy": 0.9451989244861687,
-      "eval_f1": 0.641419412554048,
-      "eval_loss": 0.14974650740623474,
-      "eval_precision": 0.7779385171790235,
-      "eval_recall": 0.545662100456621,
-      "eval_runtime": 153.6101,
-      "eval_samples_per_second": 285.697,
-      "eval_steps_per_second": 35.714,
+      "epoch": 0.75,
+      "eval_accuracy": 0.9573023866871031,
+      "eval_f1": 0.781074578989575,
+      "eval_loss": 0.11586691439151764,
+      "eval_precision": 0.8265444670739986,
+      "eval_recall": 0.7403466099118273,
+      "eval_runtime": 209.7848,
+      "eval_samples_per_second": 152.39,
+      "eval_steps_per_second": 19.053,
       "step": 2000
     },
     {
-      "epoch": 0.66,
-      "learning_rate": 1.6000000000000001e-06,
-      "loss": 0.1482,
+      "epoch": 0.79,
+      "learning_rate": 1.74e-05,
+      "loss": 0.1129,
+      "step": 2100
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.9564578185116832,
+      "eval_f1": 0.7782733354571519,
+      "eval_loss": 0.12270316481590271,
+      "eval_precision": 0.8173302107728337,
+      "eval_recall": 0.7427789601702646,
+      "eval_runtime": 209.6077,
+      "eval_samples_per_second": 152.518,
+      "eval_steps_per_second": 19.069,
       "step": 2100
     },
     {
-      "epoch": 0.69,
-      "learning_rate": 1.2000000000000002e-06,
-      "loss": 0.1449,
+      "epoch": 0.83,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.1244,
       "step": 2200
     },
     {
-      "epoch": 0.71,
-      "eval_accuracy": 0.9451989244861687,
-      "eval_f1": 0.6542056074766355,
-      "eval_loss": 0.1456586718559265,
-      "eval_precision": 0.7550614005974112,
-      "eval_recall": 0.5771182141045155,
-      "eval_runtime": 152.2253,
-      "eval_samples_per_second": 288.296,
-      "eval_steps_per_second": 36.039,
-      "step": 2250
+      "epoch": 0.83,
+      "eval_accuracy": 0.9576777503206231,
+      "eval_f1": 0.776769509981851,
+      "eval_loss": 0.11879286915063858,
+      "eval_precision": 0.8492063492063492,
+      "eval_recall": 0.7157190635451505,
+      "eval_runtime": 209.8694,
+      "eval_samples_per_second": 152.328,
+      "eval_steps_per_second": 19.045,
+      "step": 2200
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.62e-05,
+      "loss": 0.1198,
+      "step": 2300
+    },
+    {
+      "epoch": 0.87,
+      "eval_accuracy": 0.9573336669898964,
+      "eval_f1": 0.7855345911949686,
+      "eval_loss": 0.12149158120155334,
+      "eval_precision": 0.8134158254640182,
+      "eval_recall": 0.7595013681970204,
+      "eval_runtime": 209.8421,
+      "eval_samples_per_second": 152.348,
+      "eval_steps_per_second": 19.048,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.56e-05,
+      "loss": 0.1167,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9,
+      "eval_accuracy": 0.9582095154681097,
+      "eval_f1": 0.7829054273643159,
+      "eval_loss": 0.11379627883434296,
+      "eval_precision": 0.8408376963350785,
+      "eval_recall": 0.7324414715719063,
+      "eval_runtime": 210.0964,
+      "eval_samples_per_second": 152.163,
+      "eval_steps_per_second": 19.025,
+      "step": 2400
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.5e-05,
+      "loss": 0.1162,
+      "step": 2500
+    },
+    {
+      "epoch": 0.94,
+      "eval_accuracy": 0.9579592730457631,
+      "eval_f1": 0.7870722433460076,
+      "eval_loss": 0.11324156820774078,
+      "eval_precision": 0.8217002977174992,
+      "eval_recall": 0.7552447552447552,
+      "eval_runtime": 210.2617,
+      "eval_samples_per_second": 152.044,
+      "eval_steps_per_second": 19.01,
+      "step": 2500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.44e-05,
+      "loss": 0.1202,
+      "step": 2600
+    },
+    {
+      "epoch": 0.98,
+      "eval_accuracy": 0.9577403109262098,
+      "eval_f1": 0.785385226370135,
+      "eval_loss": 0.11547254770994186,
+      "eval_precision": 0.8223552894211577,
+      "eval_recall": 0.7515962298570994,
+      "eval_runtime": 209.2873,
+      "eval_samples_per_second": 152.752,
+      "eval_steps_per_second": 19.098,
+      "step": 2600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3800000000000002e-05,
+      "loss": 0.1069,
+      "step": 2700
+    },
+    {
+      "epoch": 1.02,
+      "eval_accuracy": 0.9589289624323564,
+      "eval_f1": 0.7883282282766404,
+      "eval_loss": 0.11415638774633408,
+      "eval_precision": 0.8390528483184626,
+      "eval_recall": 0.7433870477348739,
+      "eval_runtime": 209.3049,
+      "eval_samples_per_second": 152.739,
+      "eval_steps_per_second": 19.097,
+      "step": 2700
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.32e-05,
+      "loss": 0.0866,
+      "step": 2800
+    },
+    {
+      "epoch": 1.05,
+      "eval_accuracy": 0.9558947730614032,
+      "eval_f1": 0.7875866224766495,
+      "eval_loss": 0.12708334624767303,
+      "eval_precision": 0.7805315019408778,
+      "eval_recall": 0.7947704469443599,
+      "eval_runtime": 208.8539,
+      "eval_samples_per_second": 153.069,
+      "eval_steps_per_second": 19.138,
+      "step": 2800
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.26e-05,
+      "loss": 0.0913,
+      "step": 2900
+    },
+    {
+      "epoch": 1.09,
+      "eval_accuracy": 0.9557696518502299,
+      "eval_f1": 0.7855626326963907,
+      "eval_loss": 0.12602387368679047,
+      "eval_precision": 0.783661119515885,
+      "eval_recall": 0.7874733961690483,
+      "eval_runtime": 209.1412,
+      "eval_samples_per_second": 152.858,
+      "eval_steps_per_second": 19.111,
+      "step": 2900
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.2e-05,
+      "loss": 0.087,
+      "step": 3000
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.95548812912509,
+      "eval_f1": 0.7876436352783167,
+      "eval_loss": 0.12803590297698975,
+      "eval_precision": 0.7734466588511137,
+      "eval_recall": 0.8023715415019763,
+      "eval_runtime": 209.0203,
+      "eval_samples_per_second": 152.947,
+      "eval_steps_per_second": 19.123,
+      "step": 3000
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.1400000000000001e-05,
+      "loss": 0.0839,
+      "step": 3100
+    },
+    {
+      "epoch": 1.17,
+      "eval_accuracy": 0.958991523037943,
+      "eval_f1": 0.7922016167379933,
+      "eval_loss": 0.12513290345668793,
+      "eval_precision": 0.8274834437086093,
+      "eval_recall": 0.759805411979325,
+      "eval_runtime": 209.1866,
+      "eval_samples_per_second": 152.825,
+      "eval_steps_per_second": 19.107,
+      "step": 3100
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.08e-05,
+      "loss": 0.0889,
+      "step": 3200
+    },
+    {
+      "epoch": 1.2,
+      "eval_accuracy": 0.9583033563764898,
+      "eval_f1": 0.7890489001424276,
+      "eval_loss": 0.12521126866340637,
+      "eval_precision": 0.8227722772277227,
+      "eval_recall": 0.7579811492854971,
+      "eval_runtime": 209.1839,
+      "eval_samples_per_second": 152.827,
+      "eval_steps_per_second": 19.108,
+      "step": 3200
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.02e-05,
+      "loss": 0.0887,
+      "step": 3300
+    },
+    {
+      "epoch": 1.24,
+      "eval_accuracy": 0.9562701366949232,
+      "eval_f1": 0.7879890809827116,
+      "eval_loss": 0.12457986176013947,
+      "eval_precision": 0.7860816944024206,
+      "eval_recall": 0.7899057464274856,
+      "eval_runtime": 209.6152,
+      "eval_samples_per_second": 152.513,
+      "eval_steps_per_second": 19.068,
+      "step": 3300
     }
   ],
-  "max_steps": 2500,
-  "num_train_epochs": 1,
-  "total_flos": 4567250254458720.0,
+  "max_steps": 5000,
+  "num_train_epochs": 2,
+  "total_flos": 2.279247411490296e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c35cc5616b86e51cb54a61c16200f9e508724750d248536b72cc00aab9cf690c
-size 2927
+oid sha256:6b44bb80081b9b0b786dde2935ecad29da3427e1109bb02c8953fd2899a1be62
+size 2991