Training in progress, epoch 2

logs/events.out.tfevents.1709743388.1dcff7455174.7719.13

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c35d80d9ecc6e4c816ad987c1c0378c17ffad6489e0fa0fec1378e59b8e17ae0
-size 5314
+oid sha256:8557e0c5f88f5ad30a81a238602c183cc97cdb0c2046a96994fb01ab4d56dfa4
+size 5997

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c322dcb4b59626a920d30f5e070ff3482b649d4a14b4fbe008217caa06675be7
+oid sha256:fd0d6f8e320436d079f0d7746e19263f92e5a55b0adb2f3f11c8690ae59206c2
 size 17549312

run-13/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f8d22e427f88a3cccb4343fe9a81e4f810cd23d2c572e330ac5c439934459da
+oid sha256:fd0d6f8e320436d079f0d7746e19263f92e5a55b0adb2f3f11c8690ae59206c2
 size 17549312

run-13/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da2e88401839e9ab35f222ad1a67477afdcf4d64aedc15a9faec5b33140ad3ba
+oid sha256:f6e74e362199a97483614fe1679a94a8a3c83db96e104ad0fab53fe4e5f2aa6e
 size 35122746

run-13/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:827aad0fd5a1155a436d484cfb1b57ea7db209266ead7a878cac15cd726f4131
+oid sha256:1503d94ac99f2a6df5ee52252e91b9f55b6a5a1cc941650a0e87738e07d4c396
 size 1064

run-13/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8101761252446184,
+  "best_metric": 0.7954990215264188,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 3.178528308868408,
-      "learning_rate": 0.00013234646126788495,
-      "loss": 0.5867,
+      "grad_norm": 2.201279878616333,
+      "learning_rate": 0.00022445268667640978,
+      "loss": 0.5435,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7818003913894325,
-      "eval_f1": 0.8102127659574467,
-      "eval_loss": 0.47177326679229736,
-      "eval_precision": 0.7168674698795181,
-      "eval_recall": 0.9315068493150684,
-      "eval_runtime": 29.1712,
-      "eval_samples_per_second": 35.035,
-      "eval_steps_per_second": 1.097,
+      "eval_accuracy": 0.7700587084148728,
+      "eval_f1": 0.7979363714531386,
+      "eval_loss": 0.4716385006904602,
+      "eval_precision": 0.7116564417177914,
+      "eval_recall": 0.9080234833659491,
+      "eval_runtime": 24.8269,
+      "eval_samples_per_second": 41.165,
+      "eval_steps_per_second": 1.289,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 3.0951292514801025,
-      "learning_rate": 0.0,
-      "loss": 0.469,
+      "grad_norm": 3.4069817066192627,
+      "learning_rate": 0.00018704390556367485,
+      "loss": 0.4642,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8101761252446184,
-      "eval_f1": 0.830122591943958,
-      "eval_loss": 0.44528669118881226,
-      "eval_precision": 0.751188589540412,
-      "eval_recall": 0.9275929549902152,
-      "eval_runtime": 28.496,
-      "eval_samples_per_second": 35.865,
-      "eval_steps_per_second": 1.123,
+      "eval_accuracy": 0.7954990215264188,
+      "eval_f1": 0.819047619047619,
+      "eval_loss": 0.4382912516593933,
+      "eval_precision": 0.734472049689441,
+      "eval_recall": 0.9256360078277887,
+      "eval_runtime": 25.9254,
+      "eval_samples_per_second": 39.421,
+      "eval_steps_per_second": 1.234,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 192,
+  "max_steps": 672,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 7,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9936603556458727,
-    "learning_rate": 0.0002646929225357699,
-    "num_train_epochs": 2,
-    "temperature": 26
+    "alpha": 0.8348768037192943,
+    "learning_rate": 0.0002618614677891448,
+    "num_train_epochs": 7,
+    "temperature": 10
   }
 }

run-13/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:633c942d0b60f05ab3fa982f5bbaaa9bbdcf47198dfbdd5d1b639decdd30975e
+oid sha256:c39a4a57a6152c8ee182f392094f81926b25a2790f096406761336357b2e5950
 size 4920

run-13/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0ce89c0de7eff53ff8e0e099dd046cfb7cdf189abf4e221e485666006f25009
+oid sha256:ebbe2163fd46cf9e7cc5286f800c209705f03baf04ac0ebb445340b728688718
 size 17549312

run-13/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41742d4b6de56ded26c7e698ae79c2745648e36fbca57a5049fff8337116825b
+oid sha256:24e70be6e7743bdad26985d0fda97a1943f279752d9fda8ae820ca9b1e90f363
 size 35122746

run-13/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8013fe90162f1e3a3b1efc50c8ab0c3eb1c4f58fc9b4f1b892dd31ae38182f73
+oid sha256:b81c1cd831522b6845394bafe95a17be28917bfa4063ab63dab3d5be2f32d2d9
 size 1064

run-13/checkpoint-96/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7818003913894325,
+  "best_metric": 0.7700587084148728,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-96",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 3.178528308868408,
-      "learning_rate": 0.00013234646126788495,
-      "loss": 0.5867,
+      "grad_norm": 2.201279878616333,
+      "learning_rate": 0.00022445268667640978,
+      "loss": 0.5435,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7818003913894325,
-      "eval_f1": 0.8102127659574467,
-      "eval_loss": 0.47177326679229736,
-      "eval_precision": 0.7168674698795181,
-      "eval_recall": 0.9315068493150684,
-      "eval_runtime": 29.1712,
-      "eval_samples_per_second": 35.035,
-      "eval_steps_per_second": 1.097,
+      "eval_accuracy": 0.7700587084148728,
+      "eval_f1": 0.7979363714531386,
+      "eval_loss": 0.4716385006904602,
+      "eval_precision": 0.7116564417177914,
+      "eval_recall": 0.9080234833659491,
+      "eval_runtime": 24.8269,
+      "eval_samples_per_second": 41.165,
+      "eval_steps_per_second": 1.289,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 192,
+  "max_steps": 672,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 7,
   "save_steps": 500,
   "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9936603556458727,
-    "learning_rate": 0.0002646929225357699,
-    "num_train_epochs": 2,
-    "temperature": 26
+    "alpha": 0.8348768037192943,
+    "learning_rate": 0.0002618614677891448,
+    "num_train_epochs": 7,
+    "temperature": 10
   }
 }

run-13/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:633c942d0b60f05ab3fa982f5bbaaa9bbdcf47198dfbdd5d1b639decdd30975e
+oid sha256:c39a4a57a6152c8ee182f392094f81926b25a2790f096406761336357b2e5950
 size 4920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64b90b715629d1e5783934b5c99dd80382e6f66756c0da609362121ae001b883
+oid sha256:c39a4a57a6152c8ee182f392094f81926b25a2790f096406761336357b2e5950
 size 4920