Training in progress, epoch 2

logs/events.out.tfevents.1709880700.6c1c016b0e3e.7263.22

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6abe5ce1bfc1be58b6a8d72360018b4fda3dd4ff9d0508bc7e83ad4f73f621e
-size 5314
+oid sha256:6e629391b037fc2da3c18218a8b5e912b0603498e4c7ec3bfaa0fe22a4d39721
+size 5984

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:321d23ca82d0435e27df095584be4f7765ba34dcb1170dc45c03da5fa0f2852b
+oid sha256:df70405d100f7bf9dc1a1db0091dad4f85aa97b8025df5765e911bc8fa60588e
 size 17549312

run-22/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6f4518941bcca867b4097c210c9da7b27937b4f2a4d160d2ad6aa36058e2c85
+oid sha256:df70405d100f7bf9dc1a1db0091dad4f85aa97b8025df5765e911bc8fa60588e
 size 17549312

run-22/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78bc300d2ca2ff0207c7b2bf296fc7d5b707da746ac53dc0b887f90a66a35b5e
+oid sha256:931031d2f112af7462b16d2c62079551ab4c164c55f0eca71c679624702d3f65
 size 35122746

run-22/checkpoint-96/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24aa86019b8aea1c551cc1adaf38c4db2fc01de75a22af312230f6b592e0fd81
+oid sha256:6bc13e09b59a62b6fb29cad756bfeefcce6f512c71a9054cc39e336db0b532a5
 size 14054

run-22/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fe0130d07a6b0565036c3ff656cfc34f73698e07298c3d765cba0cf202b09e7
+oid sha256:ffbdb126baafe720ded4226b3309544273bca27182803f1b46020e2242d7d542
 size 1064

run-22/checkpoint-96/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.7788649706457925,
+  "best_metric": 0.8043052837573386,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-22/checkpoint-96",
-  "epoch": 1.0,
+  "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 96,
   "is_hyper_param_search": true,
@@ -10,36 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.464465618133545,
-      "learning_rate": 0.0001794635836369996,
-      "loss": 0.565,
-      "step": 96
+      "grad_norm": 1.2523858547210693,
+      "learning_rate": 0.0004199316455808347,
+      "loss": 0.5365,
+      "step": 48
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7788649706457925,
-      "eval_f1": 0.8078231292517007,
-      "eval_loss": 0.4691426157951355,
-      "eval_precision": 0.7142857142857143,
-      "eval_recall": 0.9295499021526419,
-      "eval_runtime": 25.0337,
-      "eval_samples_per_second": 40.825,
-      "eval_steps_per_second": 1.278,
+      "eval_accuracy": 0.7504892367906066,
+      "eval_f1": 0.7921760391198044,
+      "eval_loss": 0.47587937116622925,
+      "eval_precision": 0.6787709497206704,
+      "eval_recall": 0.9510763209393346,
+      "eval_runtime": 27.0541,
+      "eval_samples_per_second": 37.776,
+      "eval_steps_per_second": 0.591,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.2549729347229,
+      "learning_rate": 0.0002799544303872231,
+      "loss": 0.4645,
+      "step": 96
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8043052837573386,
+      "eval_f1": 0.8098859315589353,
+      "eval_loss": 0.4279438853263855,
+      "eval_precision": 0.7874306839186691,
+      "eval_recall": 0.8336594911937377,
+      "eval_runtime": 27.154,
+      "eval_samples_per_second": 37.637,
+      "eval_steps_per_second": 0.589,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 864,
+  "max_steps": 192,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 9,
+  "num_train_epochs": 4,
   "save_steps": 500,
-  "total_flos": 235695197280.0,
-  "train_batch_size": 32,
+  "total_flos": 471390394560.0,
+  "train_batch_size": 64,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8956145079239534,
-    "learning_rate": 0.00020189653159162458,
-    "num_train_epochs": 9,
-    "temperature": 8
+    "alpha": 0.8212346430734099,
+    "learning_rate": 0.0005599088607744462,
+    "num_train_epochs": 4,
+    "temperature": 30
   }
 }

run-22/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a59435df03122e4e47861365c1b2756f99313581e8b38e82a031f16aa354557d
+oid sha256:1a3368b5b0af1d21214eacb240625f81c73bbf2fea6c8e26456256938e253883
 size 4920