Training in progress, epoch 2

logs/events.out.tfevents.1709876067.6c1c016b0e3e.7263.11

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e6d98aed345117a010060b30bf0c50ede7b829e269a95a23a3f2d664634f193
-size 5314
+oid sha256:01074bc52c140385b20827d95ac99b468463f9584bede1a73c48a20c3a214266
+size 5984

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:361284d6cbbbf0747d1d2496b9712460a833e47055619c1ce3d78588bf871550
+oid sha256:6e11add93b37646b53de44334884f87cfb9c0a7c95ac4d83cc9b3ffce33899bd
 size 17549312

run-11/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:136877ee651c62ba812d0f99f96161f0f3ecf647b0689286df48d749d4976c6b
+oid sha256:6e11add93b37646b53de44334884f87cfb9c0a7c95ac4d83cc9b3ffce33899bd
 size 17549312

run-11/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:161aac7bcb6524ec9d72a681761a2576fed030a34ed7d6606ea3214ed68f32ae
+oid sha256:605e534cfde92b7c0c50efa96a9f6ba944650bf42f08a44ab81cca6b096fb3fa
 size 35122746

run-11/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-11/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90937cac39589ae65d6cafbc79eb7a81e4c6f2ab317854f1cdbef198f243a770
+oid sha256:3d0c495b8f481c81a3ed7d3d68c42b607133aa0e8efa74714f0b151ec15228ff
 size 1064

run-11/checkpoint-96/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.7514677103718199,
+  "best_metric": 0.8199608610567515,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/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": 8.867341995239258,
-      "learning_rate": 0.00047267095070335774,
-      "loss": 0.5752,
-      "step": 96
+      "grad_norm": 2.152926206588745,
+      "learning_rate": 0.0007816128679887146,
+      "loss": 0.591,
+      "step": 48
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7514677103718199,
-      "eval_f1": 0.765249537892791,
-      "eval_loss": 0.501445472240448,
-      "eval_precision": 0.7250437828371279,
-      "eval_recall": 0.8101761252446184,
-      "eval_runtime": 30.0696,
-      "eval_samples_per_second": 33.988,
-      "eval_steps_per_second": 1.064,
+      "eval_accuracy": 0.7524461839530333,
+      "eval_f1": 0.7944760357432982,
+      "eval_loss": 0.4880000948905945,
+      "eval_precision": 0.6791666666666667,
+      "eval_recall": 0.9569471624266145,
+      "eval_runtime": 29.551,
+      "eval_samples_per_second": 34.584,
+      "eval_steps_per_second": 0.541,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.895570278167725,
+      "learning_rate": 0.0006513440566572622,
+      "loss": 0.4603,
+      "step": 96
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8199608610567515,
+      "eval_f1": 0.8286778398510243,
+      "eval_loss": 0.4088345766067505,
+      "eval_precision": 0.7904085257548845,
+      "eval_recall": 0.8708414872798435,
+      "eval_runtime": 30.3763,
+      "eval_samples_per_second": 33.645,
+      "eval_steps_per_second": 0.527,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 192,
+  "max_steps": 336,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 7,
   "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": 0.9981740599737018,
-    "learning_rate": 0.0009453419014067155,
-    "num_train_epochs": 2,
-    "temperature": 30
+    "alpha": 0.9971844266637861,
+    "learning_rate": 0.0009118816793201671,
+    "num_train_epochs": 7,
+    "temperature": 16
   }
 }

run-11/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7ad2a3b631ed3da14471b5ac8c719b5c11baabf1e37c311322e42839c08f1d3
+oid sha256:8a283f7ca8fa712756294b7fd6d3247bcb44ee70c9bb9ae961204f34de91039c
 size 4920