training state at step 25

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.27586206896551724,
+  "epoch": 0.3448275862068966,
   "eval_steps": 1,
-  "global_step": 20,
+  "global_step": 25,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -292,6 +292,81 @@
       "eval_samples_per_second": 1.132,
       "eval_steps_per_second": 0.566,
       "step": 19
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 2.390448570251465,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 1.7181,
+      "step": 20
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "eval_loss": 1.126497507095337,
+      "eval_runtime": 17.9736,
+      "eval_samples_per_second": 1.113,
+      "eval_steps_per_second": 0.556,
+      "step": 20
+    },
+    {
+      "epoch": 0.2896551724137931,
+      "grad_norm": 2.3728342056274414,
+      "learning_rate": 4.772727272727273e-05,
+      "loss": 1.4155,
+      "step": 21
+    },
+    {
+      "epoch": 0.2896551724137931,
+      "eval_loss": 1.09345281124115,
+      "eval_runtime": 17.8283,
+      "eval_samples_per_second": 1.122,
+      "eval_steps_per_second": 0.561,
+      "step": 21
+    },
+    {
+      "epoch": 0.30344827586206896,
+      "grad_norm": 2.4872097969055176,
+      "learning_rate": 5e-05,
+      "loss": 1.3752,
+      "step": 22
+    },
+    {
+      "epoch": 0.30344827586206896,
+      "eval_loss": 1.0705276727676392,
+      "eval_runtime": 17.6481,
+      "eval_samples_per_second": 1.133,
+      "eval_steps_per_second": 0.567,
+      "step": 22
+    },
+    {
+      "epoch": 0.31724137931034485,
+      "grad_norm": 2.953234910964966,
+      "learning_rate": 4.999672209164081e-05,
+      "loss": 1.4449,
+      "step": 23
+    },
+    {
+      "epoch": 0.31724137931034485,
+      "eval_loss": 1.0468412637710571,
+      "eval_runtime": 17.6053,
+      "eval_samples_per_second": 1.136,
+      "eval_steps_per_second": 0.568,
+      "step": 23
+    },
+    {
+      "epoch": 0.3310344827586207,
+      "grad_norm": 2.47603702545166,
+      "learning_rate": 4.998688922613788e-05,
+      "loss": 1.4286,
+      "step": 24
+    },
+    {
+      "epoch": 0.3310344827586207,
+      "eval_loss": 1.0292497873306274,
+      "eval_runtime": 17.5777,
+      "eval_samples_per_second": 1.138,
+      "eval_steps_per_second": 0.569,
+      "step": 24
     }
   ],
   "logging_steps": 1,
@@ -311,7 +386,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.4706263096016896e+16,
+  "total_flos": 3.128103443541197e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null