Training in progress, step 30

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 3.2987,
     "eval_samples_per_second": 40.319,
     "eval_steps_per_second": 5.154,
-    "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.15786233500522726,
-    "train_runtime": 149.2116,
-    "train_samples_per_second": 27.719,
-    "train_steps_per_second": 0.456
+    "total_flos": 3.254692734332928e+17,
+    "train_loss": 0.01579295479071637,
+    "train_runtime": 122.9749,
+    "train_samples_per_second": 33.633,
+    "train_steps_per_second": 1.073
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbb471982be87396397be4b0e1109a341772023833225fac2262d691333b47e4
+oid sha256:03e7a48dec3e798626006da1124a5327352069797d658fa2fa6ca28b109a6675
 size 343272234

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.15786233500522726,
-    "train_runtime": 149.2116,
-    "train_samples_per_second": 27.719,
-    "train_steps_per_second": 0.456
+    "total_flos": 3.254692734332928e+17,
+    "train_loss": 0.01579295479071637,
+    "train_runtime": 122.9749,
+    "train_samples_per_second": 33.633,
+    "train_steps_per_second": 1.073
 }

trainer_state.json

@@ -1,82 +1,142 @@
 {
-  "best_metric": 0.046887390315532684,
-  "best_model_checkpoint": "/content/drive/MyDrive/Model_folder/checkpoint-60",
+  "best_metric": 0.03015263006091118,
+  "best_model_checkpoint": "/content/drive/MyDrive/Model_folder/checkpoint-30",
   "epoch": 4.0,
   "eval_steps": 30,
-  "global_step": 68,
+  "global_step": 132,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.59,
-      "learning_rate": 0.00017058823529411766,
-      "loss": 0.5948,
+      "epoch": 0.3,
+      "learning_rate": 0.00018484848484848484,
+      "loss": 0.1596,
       "step": 10
     },
     {
-      "epoch": 1.18,
-      "learning_rate": 0.0001411764705882353,
-      "loss": 0.2063,
+      "epoch": 0.61,
+      "learning_rate": 0.00016969696969696972,
+      "loss": 0.0748,
       "step": 20
     },
     {
-      "epoch": 1.76,
-      "learning_rate": 0.00011176470588235294,
-      "loss": 0.1126,
+      "epoch": 0.91,
+      "learning_rate": 0.00015454545454545454,
+      "loss": 0.0958,
       "step": 30
     },
     {
-      "epoch": 1.76,
-      "eval_loss": 0.09689675271511078,
-      "eval_matthews_correlation": 0.9667292567110256,
-      "eval_runtime": 2.7626,
-      "eval_samples_per_second": 48.143,
-      "eval_steps_per_second": 6.154,
+      "epoch": 0.91,
+      "eval_loss": 0.03015263006091118,
+      "eval_matthews_correlation": 0.9888040854737966,
+      "eval_runtime": 2.9483,
+      "eval_samples_per_second": 45.11,
+      "eval_steps_per_second": 5.766,
       "step": 30
     },
     {
-      "epoch": 2.35,
-      "learning_rate": 8.23529411764706e-05,
-      "loss": 0.0605,
+      "epoch": 1.21,
+      "learning_rate": 0.0001393939393939394,
+      "loss": 0.0802,
       "step": 40
     },
     {
-      "epoch": 2.94,
-      "learning_rate": 5.294117647058824e-05,
-      "loss": 0.0453,
+      "epoch": 1.52,
+      "learning_rate": 0.00012424242424242425,
+      "loss": 0.0318,
       "step": 50
     },
     {
-      "epoch": 3.53,
-      "learning_rate": 2.3529411764705884e-05,
-      "loss": 0.031,
+      "epoch": 1.82,
+      "learning_rate": 0.00010909090909090909,
+      "loss": 0.0289,
       "step": 60
     },
     {
-      "epoch": 3.53,
-      "eval_loss": 0.046887390315532684,
-      "eval_matthews_correlation": 0.9888040854737966,
-      "eval_runtime": 2.3092,
-      "eval_samples_per_second": 57.596,
-      "eval_steps_per_second": 7.362,
+      "epoch": 1.82,
+      "eval_loss": 0.045789625495672226,
+      "eval_matthews_correlation": 0.9888050243347044,
+      "eval_runtime": 1.9847,
+      "eval_samples_per_second": 67.012,
+      "eval_steps_per_second": 8.565,
       "step": 60
     },
+    {
+      "epoch": 2.12,
+      "learning_rate": 9.393939393939395e-05,
+      "loss": 0.0211,
+      "step": 70
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 7.878787878787879e-05,
+      "loss": 0.0097,
+      "step": 80
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 0.0085,
+      "step": 90
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 0.05030398443341255,
+      "eval_matthews_correlation": 0.9888040854737966,
+      "eval_runtime": 2.13,
+      "eval_samples_per_second": 62.441,
+      "eval_steps_per_second": 7.981,
+      "step": 90
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 4.848484848484849e-05,
+      "loss": 0.0072,
+      "step": 100
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0071,
+      "step": 110
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 0.0065,
+      "step": 120
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 0.05042246729135513,
+      "eval_matthews_correlation": 0.9888040854737966,
+      "eval_runtime": 2.8921,
+      "eval_samples_per_second": 45.988,
+      "eval_steps_per_second": 5.878,
+      "step": 120
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 3.0303030303030305e-06,
+      "loss": 0.0062,
+      "step": 130
+    },
     {
       "epoch": 4.0,
-      "step": 68,
-      "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.15786233500522726,
-      "train_runtime": 149.2116,
-      "train_samples_per_second": 27.719,
-      "train_steps_per_second": 0.456
+      "step": 132,
+      "total_flos": 3.254692734332928e+17,
+      "train_loss": 0.01579295479071637,
+      "train_runtime": 122.9749,
+      "train_samples_per_second": 33.633,
+      "train_steps_per_second": 1.073
     }
   ],
   "logging_steps": 10,
-  "max_steps": 68,
+  "max_steps": 132,
   "num_train_epochs": 4,
   "save_steps": 30,
-  "total_flos": 3.205097416476426e+17,
+  "total_flos": 3.254692734332928e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f0efe6e2503d74d756fbdf7f4557af95d22ee8da89e50aebca9c9557104d523
+oid sha256:1c4abf93990c82b841efb6a8f8aab603e55e9c850cfab05d41e1fd822a546a6d
 size 4536