End of training

trainer_state.json

@@ -3,23 +3,44 @@
   "best_model_checkpoint": null,
   "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 30,
+  "global_step": 1626,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.922509225092251,
+      "grad_norm": 0.6193047165870667,
+      "learning_rate": 3.4624846248462485e-05,
+      "loss": 0.5669,
+      "step": 500
+    },
+    {
+      "epoch": 1.8450184501845017,
+      "grad_norm": 0.5879950523376465,
+      "learning_rate": 1.924969249692497e-05,
+      "loss": 0.4066,
+      "step": 1000
+    },
+    {
+      "epoch": 2.767527675276753,
+      "grad_norm": 0.9034110307693481,
+      "learning_rate": 3.874538745387454e-06,
+      "loss": 0.3173,
+      "step": 1500
+    },
     {
       "epoch": 3.0,
-      "step": 30,
-      "total_flos": 388808857313280.0,
-      "train_loss": 1.1855101267496744,
-      "train_runtime": 71.2626,
-      "train_samples_per_second": 0.421,
-      "train_steps_per_second": 0.421
+      "step": 1626,
+      "total_flos": 2.068561910815949e+16,
+      "train_loss": 0.41841167248189814,
+      "train_runtime": 3228.8428,
+      "train_samples_per_second": 0.504,
+      "train_steps_per_second": 0.504
     }
   ],
   "logging_steps": 500,
-  "max_steps": 30,
+  "max_steps": 1626,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -35,7 +56,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 388808857313280.0,
+  "total_flos": 2.068561910815949e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null