Upload trainer_state.json

trainer_state.json

@@ -9,188 +9,188 @@
   "log_history": [
     {
       "epoch": 1.07,
-      "eval_accuracy": 0.8663101604278075,
-      "eval_loss": 0.5277199149131775,
-      "eval_recall": 0.6122448979591837,
-      "eval_runtime": 10.117,
-      "eval_samples_per_second": 18.484,
-      "eval_steps_per_second": 4.646,
+      "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,
       "step": 200
     },
     {
       "epoch": 2.14,
-      "eval_accuracy": 0.7647058823529411,
-      "eval_loss": 0.9699290990829468,
-      "eval_recall": 0.8979591836734694,
-      "eval_runtime": 10.119,
-      "eval_samples_per_second": 18.48,
-      "eval_steps_per_second": 4.645,
+      "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,
       "step": 400
     },
     {
       "epoch": 2.67,
       "learning_rate": 4.554367201426025e-05,
-      "loss": 0.4246,
+      "loss": 1.2205,
       "step": 500
     },
     {
       "epoch": 3.21,
-      "eval_accuracy": 0.8663101604278075,
-      "eval_loss": 0.7973827719688416,
-      "eval_recall": 0.5510204081632653,
-      "eval_runtime": 10.148,
-      "eval_samples_per_second": 18.427,
-      "eval_steps_per_second": 4.631,
+      "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,
       "step": 600
     },
     {
       "epoch": 4.28,
-      "eval_accuracy": 0.8716577540106952,
-      "eval_loss": 0.9027533531188965,
-      "eval_recall": 0.5714285714285714,
-      "eval_runtime": 10.102,
-      "eval_samples_per_second": 18.511,
-      "eval_steps_per_second": 4.653,
+      "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,
       "step": 800
     },
     {
       "epoch": 5.35,
       "learning_rate": 4.10873440285205e-05,
-      "loss": 0.1059,
+      "loss": 0.8231,
       "step": 1000
     },
     {
       "epoch": 5.35,
-      "eval_accuracy": 0.8609625668449198,
-      "eval_loss": 0.8567710518836975,
-      "eval_recall": 0.7346938775510204,
-      "eval_runtime": 10.117,
-      "eval_samples_per_second": 18.484,
-      "eval_steps_per_second": 4.646,
+      "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,
       "step": 1000
     },
     {
       "epoch": 6.42,
-      "eval_accuracy": 0.9037433155080213,
-      "eval_loss": 0.8681638240814209,
-      "eval_recall": 0.7755102040816326,
-      "eval_runtime": 10.157,
-      "eval_samples_per_second": 18.411,
-      "eval_steps_per_second": 4.627,
+      "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,
       "step": 1200
     },
     {
       "epoch": 7.49,
-      "eval_accuracy": 0.8823529411764706,
-      "eval_loss": 0.9168533682823181,
-      "eval_recall": 0.6938775510204082,
-      "eval_runtime": 10.271,
-      "eval_samples_per_second": 18.207,
-      "eval_steps_per_second": 4.576,
+      "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,
       "step": 1400
     },
     {
       "epoch": 8.02,
       "learning_rate": 3.6631016042780753e-05,
-      "loss": 0.0555,
+      "loss": 0.5043,
       "step": 1500
     },
     {
       "epoch": 8.56,
-      "eval_accuracy": 0.8556149732620321,
-      "eval_loss": 0.9982467293739319,
-      "eval_recall": 0.7346938775510204,
-      "eval_runtime": 10.134,
-      "eval_samples_per_second": 18.453,
-      "eval_steps_per_second": 4.638,
+      "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,
       "step": 1600
     },
     {
       "epoch": 9.63,
-      "eval_accuracy": 0.8663101604278075,
-      "eval_loss": 1.04860258102417,
-      "eval_recall": 0.7142857142857143,
-      "eval_runtime": 10.225,
-      "eval_samples_per_second": 18.289,
-      "eval_steps_per_second": 4.597,
+      "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,
       "step": 1800
     },
     {
       "epoch": 10.7,
       "learning_rate": 3.2174688057041004e-05,
-      "loss": 0.0224,
+      "loss": 0.2426,
       "step": 2000
     },
     {
       "epoch": 10.7,
-      "eval_accuracy": 0.8770053475935828,
-      "eval_loss": 1.0292185544967651,
-      "eval_recall": 0.673469387755102,
-      "eval_runtime": 10.362,
-      "eval_samples_per_second": 18.047,
-      "eval_steps_per_second": 4.536,
+      "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,
       "step": 2000
     },
     {
       "epoch": 11.76,
-      "eval_accuracy": 0.8663101604278075,
-      "eval_loss": 1.1402537822723389,
-      "eval_recall": 0.673469387755102,
-      "eval_runtime": 10.233,
-      "eval_samples_per_second": 18.274,
-      "eval_steps_per_second": 4.593,
+      "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,
       "step": 2200
     },
     {
       "epoch": 12.83,
-      "eval_accuracy": 0.8716577540106952,
-      "eval_loss": 1.1431100368499756,
-      "eval_recall": 0.673469387755102,
-      "eval_runtime": 10.226,
-      "eval_samples_per_second": 18.287,
-      "eval_steps_per_second": 4.596,
+      "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,
       "step": 2400
     },
     {
       "epoch": 13.37,
       "learning_rate": 2.7718360071301247e-05,
-      "loss": 0.0001,
+      "loss": 0.1709,
       "step": 2500
     },
     {
       "epoch": 13.9,
-      "eval_accuracy": 0.8770053475935828,
-      "eval_loss": 1.1465816497802734,
-      "eval_recall": 0.6530612244897959,
-      "eval_runtime": 10.254,
-      "eval_samples_per_second": 18.237,
-      "eval_steps_per_second": 4.584,
+      "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,
       "step": 2600
     },
     {
       "epoch": 14.97,
-      "eval_accuracy": 0.8716577540106952,
-      "eval_loss": 1.1757899522781372,
-      "eval_recall": 0.6530612244897959,
-      "eval_runtime": 10.346,
-      "eval_samples_per_second": 18.075,
-      "eval_steps_per_second": 4.543,
+      "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,
       "step": 2800
     },
     {
       "epoch": 16.04,
       "learning_rate": 2.32620320855615e-05,
-      "loss": 0.0,
+      "loss": 0.1114,
       "step": 3000
     },
     {
       "epoch": 16.04,
-      "eval_accuracy": 0.8823529411764706,
-      "eval_loss": 1.1856350898742676,
-      "eval_recall": 0.6530612244897959,
-      "eval_runtime": 10.397,
-      "eval_samples_per_second": 17.986,
-      "eval_steps_per_second": 4.521,
+      "eval_accuracy": 0.6898395721925134,
+      "eval_loss": 2.3098907470703125,
+      "eval_recall": 0.6898395721925134,
+      "eval_runtime": 10.233,
+      "eval_samples_per_second": 18.274,
+      "eval_steps_per_second": 4.593,
       "step": 3000
     }
   ],