End of training

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.3106227575602255,
-    "eval_loss": 2.317716598510742,
-    "eval_runtime": 99.9088,
-    "eval_samples": 17200,
-    "eval_samples_per_second": 172.157,
-    "eval_steps_per_second": 2.692,
-    "perplexity": 10.152465663617022,
-    "train_loss": 5.0763140617684925,
-    "train_runtime": 1146.6501,
-    "train_samples": 17053,
-    "train_samples_per_second": 44.616,
-    "train_steps_per_second": 0.699
+    "eval_accuracy": 0.731570151719994,
+    "eval_loss": 0.9372276663780212,
+    "eval_runtime": 264.4068,
+    "eval_samples": 49327,
+    "eval_samples_per_second": 186.557,
+    "eval_steps_per_second": 2.916,
+    "perplexity": 2.5528941244824197,
+    "train_loss": 2.219597038788229,
+    "train_runtime": 7937.3216,
+    "train_samples": 147965,
+    "train_samples_per_second": 55.925,
+    "train_steps_per_second": 0.874
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.3106227575602255,
-    "eval_loss": 2.317716598510742,
-    "eval_runtime": 99.9088,
-    "eval_samples": 17200,
-    "eval_samples_per_second": 172.157,
-    "eval_steps_per_second": 2.692,
-    "perplexity": 10.152465663617022
+    "eval_accuracy": 0.731570151719994,
+    "eval_loss": 0.9372276663780212,
+    "eval_runtime": 264.4068,
+    "eval_samples": 49327,
+    "eval_samples_per_second": 186.557,
+    "eval_steps_per_second": 2.916,
+    "perplexity": 2.5528941244824197
 }

runs/Apr27_08-13-09_98bd7713ce50/events.out.tfevents.1682591630.98bd7713ce50.11648.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a153e12410dcdb490ae4d0d5bd526efa96cce1b208f5196ecc6cd6c2a9f37488
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 5.0763140617684925,
-    "train_runtime": 1146.6501,
-    "train_samples": 17053,
-    "train_samples_per_second": 44.616,
-    "train_steps_per_second": 0.699
+    "train_loss": 2.219597038788229,
+    "train_runtime": 7937.3216,
+    "train_samples": 147965,
+    "train_samples_per_second": 55.925,
+    "train_steps_per_second": 0.874
 }

trainer_state.json

@@ -2,30 +2,102 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 3.0,
-  "global_step": 801,
+  "global_step": 6936,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.87,
-      "learning_rate": 1.8789013732833958e-05,
-      "loss": 6.3233,
+      "epoch": 0.22,
+      "learning_rate": 4.639561707035756e-05,
+      "loss": 6.8529,
       "step": 500
     },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.2791234140715114e-05,
+      "loss": 3.542,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.9186851211072664e-05,
+      "loss": 2.9635,
+      "step": 1500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.558246828143022e-05,
+      "loss": 2.5274,
+      "step": 2000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.1978085351787776e-05,
+      "loss": 2.2054,
+      "step": 2500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.8373702422145332e-05,
+      "loss": 1.9297,
+      "step": 3000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 2.4769319492502884e-05,
+      "loss": 1.7394,
+      "step": 3500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.116493656286044e-05,
+      "loss": 1.5874,
+      "step": 4000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.7560553633217993e-05,
+      "loss": 1.4676,
+      "step": 4500
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.395617070357555e-05,
+      "loss": 1.3606,
+      "step": 5000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.0351787773933102e-05,
+      "loss": 1.2739,
+      "step": 5500
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 6.747404844290659e-06,
+      "loss": 1.2049,
+      "step": 6000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 3.143021914648212e-06,
+      "loss": 1.1527,
+      "step": 6500
+    },
     {
       "epoch": 3.0,
-      "step": 801,
-      "total_flos": 2940962095055544.0,
-      "train_loss": 5.0763140617684925,
-      "train_runtime": 1146.6501,
-      "train_samples_per_second": 44.616,
-      "train_steps_per_second": 0.699
+      "step": 6936,
+      "total_flos": 1.9168865726431344e+16,
+      "train_loss": 2.219597038788229,
+      "train_runtime": 7937.3216,
+      "train_samples_per_second": 55.925,
+      "train_steps_per_second": 0.874
     }
   ],
-  "max_steps": 801,
+  "max_steps": 6936,
   "num_train_epochs": 3,
-  "total_flos": 2940962095055544.0,
+  "total_flos": 1.9168865726431344e+16,
   "trial_name": null,
   "trial_params": null
 }