Upload trainer_state.json with huggingface_hub

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.6527161598205566,
-  "best_model_checkpoint": "./results/checkpoint-2000",
-  "epoch": 0.9691096305269533,
+  "best_metric": 2.6806631088256836,
+  "best_model_checkpoint": "./results/checkpoint-1500",
+  "epoch": 0.726832222895215,
   "eval_steps": 250,
-  "global_step": 2000,
+  "global_step": 1500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -160,57 +160,6 @@
       "eval_samples_per_second": 47.975,
       "eval_steps_per_second": 1.5,
       "step": 1500
-    },
-    {
-      "epoch": 0.7752877044215627,
-      "grad_norm": 0.302734375,
-      "learning_rate": 0.0002431524046570296,
-      "loss": 2.6635,
-      "step": 1600
-    },
-    {
-      "epoch": 0.8237431859479104,
-      "grad_norm": 0.33203125,
-      "learning_rate": 0.00015193378712876149,
-      "loss": 2.6561,
-      "step": 1700
-    },
-    {
-      "epoch": 0.8479709267110842,
-      "eval_loss": 2.655076503753662,
-      "eval_runtime": 1225.3417,
-      "eval_samples_per_second": 47.906,
-      "eval_steps_per_second": 1.498,
-      "step": 1750
-    },
-    {
-      "epoch": 0.872198667474258,
-      "grad_norm": 0.298828125,
-      "learning_rate": 8.074887282213439e-05,
-      "loss": 2.6481,
-      "step": 1800
-    },
-    {
-      "epoch": 0.9206541490006057,
-      "grad_norm": 0.26953125,
-      "learning_rate": 3.127924912435132e-05,
-      "loss": 2.6565,
-      "step": 1900
-    },
-    {
-      "epoch": 0.9691096305269533,
-      "grad_norm": 0.416015625,
-      "learning_rate": 4.693527335575154e-06,
-      "loss": 2.6513,
-      "step": 2000
-    },
-    {
-      "epoch": 0.9691096305269533,
-      "eval_loss": 2.6527161598205566,
-      "eval_runtime": 1223.4873,
-      "eval_samples_per_second": 47.978,
-      "eval_steps_per_second": 1.5,
-      "step": 2000
     }
   ],
   "logging_steps": 100,
@@ -230,7 +179,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.678589479911424e+17,
+  "total_flos": 3.508942109933568e+17,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null