Training in progress, epoch 2

logs/events.out.tfevents.1709741585.1dcff7455174.7719.8

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a82c0bed5664814568d79fcff8242facbea97bedf9501d81f53e07d6b88ed68
-size 5314
+oid sha256:19c9bc913b8fa6bedf0b30430ed5a1abf15a6d23edfe4ededfc8f0f27caf3bd0
+size 5997

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be50a619d2660815aba11401780fcc569eb9a6f6466daae0185bc94ab79b4f8c
+oid sha256:86ed10b4586775989b9ec7f56d78779f5f14e7f453d9a572abb85dd1ca491aa0
 size 17549312

run-8/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99ca2a3e45e9c858405620a1745700d99492ed2b033c3dac18d0dc2c8b1ebab0
+oid sha256:86ed10b4586775989b9ec7f56d78779f5f14e7f453d9a572abb85dd1ca491aa0
 size 17549312

run-8/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0eceff6fd53b84d4b63fcdcc4f293b3d8154498fd0ec3d14dda3b12b916997f
+oid sha256:521092b173de47a460614f69a3e294a73e808f254a5be570100d8ff6714e4be4
 size 35122746

run-8/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea08e534f18fd263b113f672bded131169bdbe834a0c2df01d014229d1bf082e
+oid sha256:99c984ced558cfcb63a53244da63335ee7547dbfbccd2b56c8753d87e2d5cc63
 size 1064

run-8/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8199608610567515,
+  "best_metric": 0.7045009784735812,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-8/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 3.2297720909118652,
-      "learning_rate": 0.0003356185870926135,
-      "loss": 0.5529,
+      "grad_norm": 0.6551002860069275,
+      "learning_rate": 6.801658733829082e-05,
+      "loss": 0.4365,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7602739726027398,
-      "eval_f1": 0.7973531844499586,
-      "eval_loss": 0.4773445129394531,
-      "eval_precision": 0.6905444126074498,
-      "eval_recall": 0.9432485322896281,
-      "eval_runtime": 28.1534,
-      "eval_samples_per_second": 36.301,
-      "eval_steps_per_second": 1.137,
+      "eval_accuracy": 0.662426614481409,
+      "eval_f1": 0.558258642765685,
+      "eval_loss": 0.3849833011627197,
+      "eval_precision": 0.8074074074074075,
+      "eval_recall": 0.42661448140900193,
+      "eval_runtime": 25.2796,
+      "eval_samples_per_second": 40.428,
+      "eval_steps_per_second": 1.266,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.981153130531311,
-      "learning_rate": 0.00026849486967409075,
-      "loss": 0.4542,
+      "grad_norm": 1.109588384628296,
+      "learning_rate": 5.9514513921004466e-05,
+      "loss": 0.3828,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8199608610567515,
-      "eval_f1": 0.8365896980461812,
-      "eval_loss": 0.42821842432022095,
-      "eval_precision": 0.7658536585365854,
-      "eval_recall": 0.9217221135029354,
-      "eval_runtime": 28.2265,
-      "eval_samples_per_second": 36.207,
-      "eval_steps_per_second": 1.134,
+      "eval_accuracy": 0.7045009784735812,
+      "eval_f1": 0.6544622425629291,
+      "eval_loss": 0.3611763119697571,
+      "eval_precision": 0.7878787878787878,
+      "eval_recall": 0.5596868884540117,
+      "eval_runtime": 25.4122,
+      "eval_samples_per_second": 40.217,
+      "eval_steps_per_second": 1.259,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 576,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8957226795607325,
-    "learning_rate": 0.0004027423045111362,
-    "num_train_epochs": 6,
-    "temperature": 3
+    "alpha": 0.5720238366992816,
+    "learning_rate": 7.651866075557717e-05,
+    "num_train_epochs": 9,
+    "temperature": 25
   }
 }

run-8/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c62ca979c251b24426f83ee19cf958429aed20a5e41a6549bd0708ed8286fb21
+oid sha256:57c01c47cfb7f7d8530acb1f2f7c1e65992d8943357bd77df5af6cdccc8074cf
 size 4920