End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.7379204892966361,
+    "eval_loss": 0.5417497754096985,
+    "eval_runtime": 18.5026,
+    "eval_samples": 3270,
+    "eval_samples_per_second": 176.732,
+    "eval_steps_per_second": 22.105,
+    "train_loss": 0.5471283062433793,
+    "train_runtime": 1069.2073,
+    "train_samples": 9427,
+    "train_samples_per_second": 44.084,
+    "train_steps_per_second": 1.38
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.7379204892966361,
+    "eval_loss": 0.5417497754096985,
+    "eval_runtime": 18.5026,
+    "eval_samples": 3270,
+    "eval_samples_per_second": 176.732,
+    "eval_steps_per_second": 22.105
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.5471283062433793,
+    "train_runtime": 1069.2073,
+    "train_samples": 9427,
+    "train_samples_per_second": 44.084,
+    "train_steps_per_second": 1.38
+}

trainer_state.json

@@ -0,0 +1,82 @@
+{
+  "best_metric": 0.5417497754096985,
+  "best_model_checkpoint": "./MiniLMv2-L6-H768-distilled-from-RoBERTa-Large_boolq/checkpoint-1250",
+  "epoch": 5.0,
+  "global_step": 1475,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.85,
+      "eval_accuracy": 0.618960244648318,
+      "eval_loss": 0.6579468846321106,
+      "eval_runtime": 18.5057,
+      "eval_samples_per_second": 176.702,
+      "eval_steps_per_second": 22.101,
+      "step": 250
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 6.610169491525424e-06,
+      "loss": 0.6352,
+      "step": 500
+    },
+    {
+      "epoch": 1.69,
+      "eval_accuracy": 0.6840978593272171,
+      "eval_loss": 0.5906558632850647,
+      "eval_runtime": 18.6795,
+      "eval_samples_per_second": 175.058,
+      "eval_steps_per_second": 21.896,
+      "step": 500
+    },
+    {
+      "epoch": 2.54,
+      "eval_accuracy": 0.7195718654434251,
+      "eval_loss": 0.5612974166870117,
+      "eval_runtime": 18.6317,
+      "eval_samples_per_second": 175.507,
+      "eval_steps_per_second": 21.952,
+      "step": 750
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 3.2203389830508473e-06,
+      "loss": 0.535,
+      "step": 1000
+    },
+    {
+      "epoch": 3.39,
+      "eval_accuracy": 0.7373088685015291,
+      "eval_loss": 0.5444376468658447,
+      "eval_runtime": 18.6673,
+      "eval_samples_per_second": 175.173,
+      "eval_steps_per_second": 21.91,
+      "step": 1000
+    },
+    {
+      "epoch": 4.24,
+      "eval_accuracy": 0.7379204892966361,
+      "eval_loss": 0.5417497754096985,
+      "eval_runtime": 18.6845,
+      "eval_samples_per_second": 175.012,
+      "eval_steps_per_second": 21.89,
+      "step": 1250
+    },
+    {
+      "epoch": 5.0,
+      "step": 1475,
+      "total_flos": 6243850835650560.0,
+      "train_loss": 0.5471283062433793,
+      "train_runtime": 1069.2073,
+      "train_samples_per_second": 44.084,
+      "train_steps_per_second": 1.38
+    }
+  ],
+  "max_steps": 1475,
+  "num_train_epochs": 5,
+  "total_flos": 6243850835650560.0,
+  "trial_name": null,
+  "trial_params": null
+}