Upload trainer_state.json

trainer_state.json

@@ -1,202 +1,304 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 16.0427807486631,
-  "global_step": 3000,
+  "epoch": 21.390374331550802,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.07,
-      "eval_accuracy": 0.5401069518716578,
-      "eval_loss": 1.1939018964767456,
-      "eval_recall": 0.5401069518716578,
-      "eval_runtime": 10.154,
-      "eval_samples_per_second": 18.416,
-      "eval_steps_per_second": 4.629,
+      "eval_accuracy": 0.48128342245989303,
+      "eval_f1": 0.48128342245989303,
+      "eval_loss": 1.350968837738037,
+      "eval_precision": 0.48128342245989303,
+      "eval_recall": 0.48128342245989303,
+      "eval_runtime": 10.797,
+      "eval_samples_per_second": 17.32,
+      "eval_steps_per_second": 4.353,
       "step": 200
     },
     {
       "epoch": 2.14,
-      "eval_accuracy": 0.6951871657754011,
-      "eval_loss": 0.8892518877983093,
-      "eval_recall": 0.6951871657754011,
-      "eval_runtime": 10.177,
-      "eval_samples_per_second": 18.375,
-      "eval_steps_per_second": 4.618,
+      "eval_accuracy": 0.6577540106951871,
+      "eval_f1": 0.6577540106951871,
+      "eval_loss": 1.0548419952392578,
+      "eval_precision": 0.6577540106951871,
+      "eval_recall": 0.6577540106951871,
+      "eval_runtime": 10.384,
+      "eval_samples_per_second": 18.008,
+      "eval_steps_per_second": 4.526,
       "step": 400
     },
     {
       "epoch": 2.67,
-      "learning_rate": 4.554367201426025e-05,
-      "loss": 1.2205,
+      "learning_rate": 4.732620320855615e-05,
+      "loss": 1.271,
       "step": 500
     },
     {
       "epoch": 3.21,
-      "eval_accuracy": 0.6149732620320856,
-      "eval_loss": 1.2934410572052002,
-      "eval_recall": 0.6149732620320856,
-      "eval_runtime": 10.164,
-      "eval_samples_per_second": 18.398,
-      "eval_steps_per_second": 4.624,
+      "eval_accuracy": 0.6631016042780749,
+      "eval_f1": 0.6631016042780749,
+      "eval_loss": 1.257256031036377,
+      "eval_precision": 0.6631016042780749,
+      "eval_recall": 0.6631016042780749,
+      "eval_runtime": 10.09,
+      "eval_samples_per_second": 18.533,
+      "eval_steps_per_second": 4.658,
       "step": 600
     },
     {
       "epoch": 4.28,
-      "eval_accuracy": 0.6470588235294118,
-      "eval_loss": 1.179315447807312,
-      "eval_recall": 0.6470588235294118,
-      "eval_runtime": 10.156,
-      "eval_samples_per_second": 18.413,
-      "eval_steps_per_second": 4.628,
+      "eval_accuracy": 0.6577540106951871,
+      "eval_f1": 0.6577540106951871,
+      "eval_loss": 1.3473643064498901,
+      "eval_precision": 0.6577540106951871,
+      "eval_recall": 0.6577540106951871,
+      "eval_runtime": 10.111,
+      "eval_samples_per_second": 18.495,
+      "eval_steps_per_second": 4.648,
       "step": 800
     },
     {
       "epoch": 5.35,
-      "learning_rate": 4.10873440285205e-05,
-      "loss": 0.8231,
+      "learning_rate": 4.4652406417112304e-05,
+      "loss": 0.791,
       "step": 1000
     },
     {
       "epoch": 5.35,
-      "eval_accuracy": 0.6310160427807486,
-      "eval_loss": 1.7959877252578735,
-      "eval_recall": 0.6310160427807486,
-      "eval_runtime": 10.15,
-      "eval_samples_per_second": 18.424,
-      "eval_steps_per_second": 4.631,
+      "eval_accuracy": 0.6844919786096256,
+      "eval_f1": 0.6844919786096256,
+      "eval_loss": 1.6341558694839478,
+      "eval_precision": 0.6844919786096256,
+      "eval_recall": 0.6844919786096256,
+      "eval_runtime": 10.125,
+      "eval_samples_per_second": 18.469,
+      "eval_steps_per_second": 4.642,
       "step": 1000
     },
     {
       "epoch": 6.42,
-      "eval_accuracy": 0.7005347593582888,
-      "eval_loss": 1.6884827613830566,
-      "eval_recall": 0.7005347593582888,
-      "eval_runtime": 10.222,
-      "eval_samples_per_second": 18.294,
-      "eval_steps_per_second": 4.598,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_f1": 0.7272727272727273,
+      "eval_loss": 1.7296054363250732,
+      "eval_precision": 0.7272727272727273,
+      "eval_recall": 0.7272727272727273,
+      "eval_runtime": 10.343,
+      "eval_samples_per_second": 18.08,
+      "eval_steps_per_second": 4.544,
       "step": 1200
     },
     {
       "epoch": 7.49,
-      "eval_accuracy": 0.6203208556149733,
-      "eval_loss": 1.8338332176208496,
-      "eval_recall": 0.6203208556149733,
-      "eval_runtime": 10.15,
-      "eval_samples_per_second": 18.424,
-      "eval_steps_per_second": 4.631,
+      "eval_accuracy": 0.7112299465240641,
+      "eval_f1": 0.7112299465240641,
+      "eval_loss": 1.687750220298767,
+      "eval_precision": 0.7112299465240641,
+      "eval_recall": 0.7112299465240641,
+      "eval_runtime": 10.345,
+      "eval_samples_per_second": 18.076,
+      "eval_steps_per_second": 4.543,
       "step": 1400
     },
     {
       "epoch": 8.02,
-      "learning_rate": 3.6631016042780753e-05,
-      "loss": 0.5043,
+      "learning_rate": 4.197860962566845e-05,
+      "loss": 0.5343,
       "step": 1500
     },
     {
       "epoch": 8.56,
-      "eval_accuracy": 0.6524064171122995,
-      "eval_loss": 1.8866851329803467,
-      "eval_recall": 0.6524064171122995,
-      "eval_runtime": 10.282,
-      "eval_samples_per_second": 18.187,
-      "eval_steps_per_second": 4.571,
+      "eval_accuracy": 0.6951871657754011,
+      "eval_f1": 0.6951871657754011,
+      "eval_loss": 1.8570679426193237,
+      "eval_precision": 0.6951871657754011,
+      "eval_recall": 0.6951871657754011,
+      "eval_runtime": 10.19,
+      "eval_samples_per_second": 18.351,
+      "eval_steps_per_second": 4.612,
       "step": 1600
     },
     {
       "epoch": 9.63,
-      "eval_accuracy": 0.6951871657754011,
-      "eval_loss": 1.6332286596298218,
-      "eval_recall": 0.6951871657754011,
-      "eval_runtime": 10.094,
-      "eval_samples_per_second": 18.526,
-      "eval_steps_per_second": 4.656,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_f1": 0.7058823529411765,
+      "eval_loss": 1.9290939569473267,
+      "eval_precision": 0.7058823529411765,
+      "eval_recall": 0.7058823529411765,
+      "eval_runtime": 10.161,
+      "eval_samples_per_second": 18.404,
+      "eval_steps_per_second": 4.626,
       "step": 1800
     },
     {
       "epoch": 10.7,
-      "learning_rate": 3.2174688057041004e-05,
-      "loss": 0.2426,
+      "learning_rate": 3.93048128342246e-05,
+      "loss": 0.2934,
       "step": 2000
     },
     {
       "epoch": 10.7,
-      "eval_accuracy": 0.6631016042780749,
-      "eval_loss": 2.0470945835113525,
-      "eval_recall": 0.6631016042780749,
-      "eval_runtime": 10.46,
-      "eval_samples_per_second": 17.878,
-      "eval_steps_per_second": 4.493,
+      "eval_accuracy": 0.7005347593582888,
+      "eval_f1": 0.7005347593582888,
+      "eval_loss": 1.9618135690689087,
+      "eval_precision": 0.7005347593582888,
+      "eval_recall": 0.7005347593582888,
+      "eval_runtime": 10.233,
+      "eval_samples_per_second": 18.274,
+      "eval_steps_per_second": 4.593,
       "step": 2000
     },
     {
       "epoch": 11.76,
-      "eval_accuracy": 0.6684491978609626,
-      "eval_loss": 2.127241373062134,
-      "eval_recall": 0.6684491978609626,
-      "eval_runtime": 10.207,
-      "eval_samples_per_second": 18.321,
-      "eval_steps_per_second": 4.605,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_f1": 0.7272727272727273,
+      "eval_loss": 1.8590352535247803,
+      "eval_precision": 0.7272727272727273,
+      "eval_recall": 0.7272727272727273,
+      "eval_runtime": 10.155,
+      "eval_samples_per_second": 18.415,
+      "eval_steps_per_second": 4.628,
       "step": 2200
     },
     {
       "epoch": 12.83,
-      "eval_accuracy": 0.7112299465240641,
-      "eval_loss": 2.163196563720703,
-      "eval_recall": 0.7112299465240641,
-      "eval_runtime": 10.211,
-      "eval_samples_per_second": 18.314,
-      "eval_steps_per_second": 4.603,
+      "eval_accuracy": 0.732620320855615,
+      "eval_f1": 0.7326203208556151,
+      "eval_loss": 2.042591094970703,
+      "eval_precision": 0.732620320855615,
+      "eval_recall": 0.732620320855615,
+      "eval_runtime": 10.107,
+      "eval_samples_per_second": 18.502,
+      "eval_steps_per_second": 4.65,
       "step": 2400
     },
     {
       "epoch": 13.37,
-      "learning_rate": 2.7718360071301247e-05,
-      "loss": 0.1709,
+      "learning_rate": 3.6631016042780753e-05,
+      "loss": 0.1892,
       "step": 2500
     },
     {
       "epoch": 13.9,
-      "eval_accuracy": 0.679144385026738,
-      "eval_loss": 2.2403364181518555,
-      "eval_recall": 0.679144385026738,
-      "eval_runtime": 10.385,
-      "eval_samples_per_second": 18.007,
-      "eval_steps_per_second": 4.526,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_f1": 0.7058823529411765,
+      "eval_loss": 2.264678716659546,
+      "eval_precision": 0.7058823529411765,
+      "eval_recall": 0.7058823529411765,
+      "eval_runtime": 10.161,
+      "eval_samples_per_second": 18.404,
+      "eval_steps_per_second": 4.626,
       "step": 2600
     },
     {
       "epoch": 14.97,
-      "eval_accuracy": 0.6844919786096256,
-      "eval_loss": 2.3301198482513428,
-      "eval_recall": 0.6844919786096256,
-      "eval_runtime": 10.223,
-      "eval_samples_per_second": 18.292,
-      "eval_steps_per_second": 4.597,
+      "eval_accuracy": 0.7112299465240641,
+      "eval_f1": 0.7112299465240641,
+      "eval_loss": 2.111647844314575,
+      "eval_precision": 0.7112299465240641,
+      "eval_recall": 0.7112299465240641,
+      "eval_runtime": 10.145,
+      "eval_samples_per_second": 18.433,
+      "eval_steps_per_second": 4.633,
       "step": 2800
     },
     {
       "epoch": 16.04,
-      "learning_rate": 2.32620320855615e-05,
-      "loss": 0.1114,
+      "learning_rate": 3.39572192513369e-05,
+      "loss": 0.129,
       "step": 3000
     },
     {
       "epoch": 16.04,
+      "eval_accuracy": 0.7165775401069518,
+      "eval_f1": 0.7165775401069518,
+      "eval_loss": 2.1683263778686523,
+      "eval_precision": 0.7165775401069518,
+      "eval_recall": 0.7165775401069518,
+      "eval_runtime": 10.287,
+      "eval_samples_per_second": 18.178,
+      "eval_steps_per_second": 4.569,
+      "step": 3000
+    },
+    {
+      "epoch": 17.11,
+      "eval_accuracy": 0.7165775401069518,
+      "eval_f1": 0.7165775401069518,
+      "eval_loss": 2.329145669937134,
+      "eval_precision": 0.7165775401069518,
+      "eval_recall": 0.7165775401069518,
+      "eval_runtime": 10.188,
+      "eval_samples_per_second": 18.355,
+      "eval_steps_per_second": 4.613,
+      "step": 3200
+    },
+    {
+      "epoch": 18.18,
       "eval_accuracy": 0.6898395721925134,
-      "eval_loss": 2.3098907470703125,
+      "eval_f1": 0.6898395721925134,
+      "eval_loss": 2.538438081741333,
+      "eval_precision": 0.6898395721925134,
       "eval_recall": 0.6898395721925134,
-      "eval_runtime": 10.233,
-      "eval_samples_per_second": 18.274,
-      "eval_steps_per_second": 4.593,
-      "step": 3000
+      "eval_runtime": 10.202,
+      "eval_samples_per_second": 18.33,
+      "eval_steps_per_second": 4.607,
+      "step": 3400
+    },
+    {
+      "epoch": 18.72,
+      "learning_rate": 3.128342245989305e-05,
+      "loss": 0.1084,
+      "step": 3500
+    },
+    {
+      "epoch": 19.25,
+      "eval_accuracy": 0.6951871657754011,
+      "eval_f1": 0.6951871657754011,
+      "eval_loss": 2.630084753036499,
+      "eval_precision": 0.6951871657754011,
+      "eval_recall": 0.6951871657754011,
+      "eval_runtime": 10.362,
+      "eval_samples_per_second": 18.047,
+      "eval_steps_per_second": 4.536,
+      "step": 3600
+    },
+    {
+      "epoch": 20.32,
+      "eval_accuracy": 0.6684491978609626,
+      "eval_f1": 0.6684491978609626,
+      "eval_loss": 2.8742926120758057,
+      "eval_precision": 0.6684491978609626,
+      "eval_recall": 0.6684491978609626,
+      "eval_runtime": 10.323,
+      "eval_samples_per_second": 18.115,
+      "eval_steps_per_second": 4.553,
+      "step": 3800
+    },
+    {
+      "epoch": 21.39,
+      "learning_rate": 2.8609625668449196e-05,
+      "loss": 0.0935,
+      "step": 4000
+    },
+    {
+      "epoch": 21.39,
+      "eval_accuracy": 0.7219251336898396,
+      "eval_f1": 0.7219251336898396,
+      "eval_loss": 2.320441484451294,
+      "eval_precision": 0.7219251336898396,
+      "eval_recall": 0.7219251336898396,
+      "eval_runtime": 10.306,
+      "eval_samples_per_second": 18.145,
+      "eval_steps_per_second": 4.56,
+      "step": 4000
     }
   ],
-  "max_steps": 5610,
-  "num_train_epochs": 30,
-  "total_flos": 3149026201829376.0,
+  "max_steps": 9350,
+  "num_train_epochs": 50,
+  "total_flos": 4198877016109056.0,
   "trial_name": null,
   "trial_params": null
 }