Upload trainer_state.json with huggingface_hub

trainer_state.json

@@ -1,45 +1,109 @@
 {
-  "best_global_step": 1846,
-  "best_metric": 0.8862317158114698,
-  "best_model_checkpoint": "outputs/bartpho-hsd/checkpoint-best/checkpoint-1846",
-  "epoch": 2.0,
+  "best_global_step": 5538,
+  "best_metric": 0.8634780589737636,
+  "best_model_checkpoint": "outputs/bartpho-hsd/checkpoint-best/checkpoint-5538",
+  "epoch": 6.0,
   "eval_steps": 500,
-  "global_step": 1846,
+  "global_step": 5538,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 7.859710216522217,
+      "grad_norm": 2.5697989463806152,
       "learning_rate": 4.998769072267837e-05,
-      "loss": 0.3322,
+      "loss": 0.5389,
       "step": 923
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.885767355467843,
-      "eval_loss": 0.3262785077095032,
-      "eval_runtime": 26.1724,
-      "eval_samples_per_second": 164.563,
-      "eval_steps_per_second": 5.158,
+      "eval_accuracy": 0.8428140236823776,
+      "eval_loss": 0.5179033875465393,
+      "eval_runtime": 60.9906,
+      "eval_samples_per_second": 70.617,
+      "eval_steps_per_second": 2.213,
       "step": 923
     },
     {
       "epoch": 2.0,
-      "grad_norm": 8.771722793579102,
+      "grad_norm": 3.148050308227539,
       "learning_rate": 4.995072162589518e-05,
-      "loss": 0.1918,
+      "loss": 0.5208,
       "step": 1846
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8862317158114698,
-      "eval_loss": 0.4059225618839264,
-      "eval_runtime": 26.1398,
-      "eval_samples_per_second": 164.768,
-      "eval_steps_per_second": 5.165,
+      "eval_accuracy": 0.8428140236823776,
+      "eval_loss": 0.5180655121803284,
+      "eval_runtime": 55.4284,
+      "eval_samples_per_second": 77.704,
+      "eval_steps_per_second": 2.436,
       "step": 1846
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 6.275445461273193,
+      "learning_rate": 4.988912917920435e-05,
+      "loss": 0.4785,
+      "step": 2769
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.850243789180404,
+      "eval_loss": 0.48658695816993713,
+      "eval_runtime": 55.6378,
+      "eval_samples_per_second": 77.411,
+      "eval_steps_per_second": 2.426,
+      "step": 2769
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.8800065517425537,
+      "learning_rate": 4.980297416691463e-05,
+      "loss": 0.4072,
+      "step": 3692
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8379382400742976,
+      "eval_loss": 0.45363402366638184,
+      "eval_runtime": 55.9448,
+      "eval_samples_per_second": 76.987,
+      "eval_steps_per_second": 2.413,
+      "step": 3692
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.930692195892334,
+      "learning_rate": 4.969234161362153e-05,
+      "loss": 0.372,
+      "step": 4615
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8576735546784304,
+      "eval_loss": 0.4497613310813904,
+      "eval_runtime": 55.8766,
+      "eval_samples_per_second": 77.081,
+      "eval_steps_per_second": 2.416,
+      "step": 4615
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.390305995941162,
+      "learning_rate": 4.9557340700298316e-05,
+      "loss": 0.345,
+      "step": 5538
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8634780589737636,
+      "eval_loss": 0.4286285638809204,
+      "eval_runtime": 55.8527,
+      "eval_samples_per_second": 77.114,
+      "eval_steps_per_second": 2.417,
+      "step": 5538
     }
   ],
   "logging_steps": 923,
@@ -68,7 +132,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7767634318166016.0,
+  "total_flos": 9.625341380307149e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null