Training in progress, epoch 1

logs/events.out.tfevents.1709696898.a2333cf40ebd.21010.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba3d1262e27239036198483f135b4eb6f59bff3e2d060ae1235749eb6630829e
-size 5314
+oid sha256:472d2c079ca368f63e847799aa435aa51a891cbb7f605f5b0397c83e12d17732
+size 6351

logs/events.out.tfevents.1709697132.a2333cf40ebd.21010.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a291ac31fba750739770fa01afde949394278bdc98f20828e00ff7238845a6ab
+size 5315

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57ef7ac15d70c06615ec62775bbd9d47a7a307f789675ebd91d014018802f0ac
+oid sha256:446c2c31d01eea9266672f75b6ffe66317a4cdf5130baa8b7dd4d988ec7f3dd9
 size 17549312

run-2/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc3ae2fa78db1d16939496673fe96513e6285b61d2f7edfaba98bf3bfef765fa
+oid sha256:444ec7032343c5e48a0bebcbc727c0d6dcc53de822b7281218d78d3d52fc74a5
 size 17549312

run-2/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e884cf9ab66245efe4e20a0e9734d0a79152a86f3e758b388219b9f831cdec1
+oid sha256:a7d65e468a5ea6ece563e4a260f8a916b8919ff2c935298cbc0e77b353da1ae9
 size 35122746

run-2/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0dfe3c3c7200f4f9083675df7ddf4d22895ef93dc1da39a03576fcc4c7d46673
+oid sha256:76b1ca96b43548704a5f4d11597321c5afbd7e065f2cc83810db95ee4c1c77f7
 size 1064

run-2/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7964774951076321,
+  "best_metric": 0.8140900195694716,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.613167643547058,
-      "learning_rate": 0.00027081825418014463,
-      "loss": 0.516,
+      "grad_norm": 3.4323604106903076,
+      "learning_rate": 0.0004667999543114754,
+      "loss": 0.533,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7837573385518591,
-      "eval_f1": 0.8073234524847429,
-      "eval_loss": 0.44109782576560974,
-      "eval_precision": 0.7279874213836478,
-      "eval_recall": 0.9060665362035225,
-      "eval_runtime": 31.7637,
-      "eval_samples_per_second": 32.175,
-      "eval_steps_per_second": 1.007,
+      "eval_accuracy": 0.7984344422700587,
+      "eval_f1": 0.8049242424242424,
+      "eval_loss": 0.4442267119884491,
+      "eval_precision": 0.7798165137614679,
+      "eval_recall": 0.8317025440313112,
+      "eval_runtime": 28.6457,
+      "eval_samples_per_second": 35.677,
+      "eval_steps_per_second": 1.117,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.826476812362671,
-      "learning_rate": 0.00013540912709007232,
-      "loss": 0.4436,
+      "grad_norm": 2.4163033962249756,
+      "learning_rate": 0.0,
+      "loss": 0.4374,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7964774951076321,
-      "eval_f1": 0.820069204152249,
-      "eval_loss": 0.42240670323371887,
-      "eval_precision": 0.7348837209302326,
-      "eval_recall": 0.9275929549902152,
-      "eval_runtime": 30.3967,
-      "eval_samples_per_second": 33.622,
-      "eval_steps_per_second": 1.053,
+      "eval_accuracy": 0.8140900195694716,
+      "eval_f1": 0.8243992606284659,
+      "eval_loss": 0.4299594759941101,
+      "eval_precision": 0.7810858143607706,
+      "eval_recall": 0.87279843444227,
+      "eval_runtime": 28.93,
+      "eval_samples_per_second": 35.327,
+      "eval_steps_per_second": 1.106,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 288,
+  "max_steps": 192,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7853139256228429,
-    "learning_rate": 0.00040622738127021695,
-    "num_train_epochs": 3,
-    "temperature": 5
+    "alpha": 0.88253398201078,
+    "learning_rate": 0.0009335999086229508,
+    "num_train_epochs": 2,
+    "temperature": 19
   }
 }

run-2/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db0762669f3cc0b738504b8520ffeb23dad59a4047901db16223a1552e162134
+oid sha256:2659042d49fa36ce1404b1f7057dd4643dc22f653d1fc44d521700862149e4b7
 size 4920

run-3/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af020de10a78437284341857ed456069cddcba5115cd9cf73bbd880c668987f4
+oid sha256:446c2c31d01eea9266672f75b6ffe66317a4cdf5130baa8b7dd4d988ec7f3dd9
 size 17549312

run-3/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a1e62db45f26fc9aceaafe4cfd47f298fe6070cc9e9df068869a51b06a179fb
+oid sha256:bde5e387217b41822b0ca9ea98caa7129bea059867adb471204d76b01650315d
 size 35122746

run-3/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73d0eae126ade6de0e19f38bc7c732485cd14e2a7620ccdfba356009a97cdf22
+oid sha256:f89cc7dd2ab1ceae4cc5d3844408884d87d0a8424a3ba21161199ee633ef9b97
 size 1064

run-3/checkpoint-96/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.786692759295499,
+  "best_metric": 0.5,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-96",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.2690824270248413,
-      "learning_rate": 0.00018636956997148543,
-      "loss": 0.4818,
+      "grad_norm": 2.101290702819824,
+      "learning_rate": 1.1962059637335337e-06,
+      "loss": 0.5173,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.786692759295499,
-      "eval_f1": 0.8039568345323741,
-      "eval_loss": 0.4252791106700897,
-      "eval_precision": 0.7437603993344426,
-      "eval_recall": 0.8747553816046967,
-      "eval_runtime": 31.9977,
-      "eval_samples_per_second": 31.94,
-      "eval_steps_per_second": 1.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.6666666666666666,
+      "eval_loss": 0.5046124458312988,
+      "eval_precision": 0.5,
+      "eval_recall": 1.0,
+      "eval_runtime": 28.251,
+      "eval_samples_per_second": 36.176,
+      "eval_steps_per_second": 1.133,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 672,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6895915406036343,
-    "learning_rate": 0.000217431164966733,
-    "num_train_epochs": 7,
-    "temperature": 20
+    "alpha": 0.586704262013931,
+    "learning_rate": 1.3457317092002255e-06,
+    "num_train_epochs": 9,
+    "temperature": 21
   }
 }

run-3/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d383e461806a90d203afa329c2e34418d5e578d8a577a59888ecbd6bee4250c7
+oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
 size 4920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2659042d49fa36ce1404b1f7057dd4643dc22f653d1fc44d521700862149e4b7
+oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
 size 4920