Training in progress, epoch 2

logs/events.out.tfevents.1711300537.73e3a81c01ef.4225.33

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:096d1be01deb6bae449c0b5274004e8ee6a82a47d6f3f9ec1cf6ff740451bc8d
-size 5406
+oid sha256:f2693cfac8456eeec90578886cd85f8194f684b1aa3feb23a967b8289cd589a5
+size 6136

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26c742618d7bc5ebfc93debcdbbcdf43a88db214cf4e3228b885ff412aa746b3
+oid sha256:3112f0d0bdf216572b425869d1615693330c99f37b8ad55d1ca3737f15e62147
 size 17549312

run-33/checkpoint-384/config.json

@@ -27,7 +27,7 @@
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.38.2",
+  "transformers_version": "4.39.1",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

run-33/checkpoint-384/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3517bab38dc549c6d9d5ada7b51bd830910fe7d5d8ef0d9d01d431ae60232428
+oid sha256:3112f0d0bdf216572b425869d1615693330c99f37b8ad55d1ca3737f15e62147
 size 17549312

run-33/checkpoint-384/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9310b523d26b0d739c57387f369d9d0bbe8a4924c62e4f045f0355b04447f7c1
+oid sha256:70b6d4a2a0ff17ab288833a0034413758323fbfbd5f061a035a6e09a0f741ca8
 size 35122746

run-33/checkpoint-384/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45b1fc07ed8b121716021ec87c686f1cd9c3b89c82ea08a6a0792d47a39077c9
+oid sha256:d2558f72cda987826e5e7caf54cc6282fe335ebecbddfed7bb83f0184d1f54cc
 size 14054

run-33/checkpoint-384/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:784c97862fadcdc04fbcd9c34f90b66c1ff159e0529c09d99e56334bf96b3ba8
+oid sha256:55b59249e23a9045246b4a07400e5db5dc6c849b6ef79aba1e160d1db99404ba
 size 1064

run-33/checkpoint-384/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.8238747553816047,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-33/checkpoint-288",
-  "epoch": 4.0,
+  "best_metric": 0.8140900195694716,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-33/checkpoint-384",
+  "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 384,
   "is_hyper_param_search": true,
@@ -10,93 +10,58 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.87579345703125,
-      "learning_rate": 0.00016441875793172356,
-      "loss": 0.5654,
-      "step": 96
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.7896281800391389,
-      "eval_f1": 0.816396242527754,
-      "eval_loss": 0.4605298936367035,
-      "eval_precision": 0.7242424242424242,
-      "eval_recall": 0.9354207436399217,
-      "eval_runtime": 29.158,
-      "eval_samples_per_second": 35.05,
-      "eval_steps_per_second": 1.097,
-      "step": 96
-    },
-    {
-      "epoch": 2.0,
-      "grad_norm": 5.452133655548096,
-      "learning_rate": 0.0001096125052878157,
-      "loss": 0.4617,
+      "grad_norm": 2.441394090652466,
+      "learning_rate": 0.00039892736546732013,
+      "loss": 0.5568,
       "step": 192
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.8101761252446184,
-      "eval_f1": 0.829225352112676,
-      "eval_loss": 0.43088027834892273,
-      "eval_precision": 0.7536,
+      "epoch": 1.0,
+      "eval_accuracy": 0.7818003913894325,
+      "eval_f1": 0.8085836909871245,
+      "eval_loss": 0.45964643359184265,
+      "eval_mcc": 0.5870562340513658,
+      "eval_precision": 0.7201834862385321,
       "eval_recall": 0.9217221135029354,
-      "eval_runtime": 29.758,
-      "eval_samples_per_second": 34.344,
-      "eval_steps_per_second": 1.075,
+      "eval_runtime": 67.0533,
+      "eval_samples_per_second": 15.242,
+      "eval_steps_per_second": 0.477,
       "step": 192
     },
     {
-      "epoch": 3.0,
-      "grad_norm": 4.057122230529785,
-      "learning_rate": 5.480625264390785e-05,
-      "loss": 0.4328,
-      "step": 288
-    },
-    {
-      "epoch": 3.0,
-      "eval_accuracy": 0.8238747553816047,
-      "eval_f1": 0.832089552238806,
-      "eval_loss": 0.4205124080181122,
-      "eval_precision": 0.7950089126559715,
-      "eval_recall": 0.87279843444227,
-      "eval_runtime": 29.1506,
-      "eval_samples_per_second": 35.059,
-      "eval_steps_per_second": 1.098,
-      "step": 288
-    },
-    {
-      "epoch": 4.0,
-      "grad_norm": 4.137439727783203,
-      "learning_rate": 0.0,
-      "loss": 0.4192,
+      "epoch": 2.0,
+      "grad_norm": 6.07141637802124,
+      "learning_rate": 0.0002659515769782134,
+      "loss": 0.4779,
       "step": 384
     },
     {
-      "epoch": 4.0,
-      "eval_accuracy": 0.8238747553816047,
-      "eval_f1": 0.8375451263537905,
-      "eval_loss": 0.41125524044036865,
-      "eval_precision": 0.7772194304857621,
-      "eval_recall": 0.9080234833659491,
-      "eval_runtime": 28.1675,
-      "eval_samples_per_second": 36.283,
-      "eval_steps_per_second": 1.136,
+      "epoch": 2.0,
+      "eval_accuracy": 0.8140900195694716,
+      "eval_f1": 0.8227611940298508,
+      "eval_loss": 0.44287753105163574,
+      "eval_mcc": 0.6312089367953639,
+      "eval_precision": 0.786096256684492,
+      "eval_recall": 0.863013698630137,
+      "eval_runtime": 66.4724,
+      "eval_samples_per_second": 15.375,
+      "eval_steps_per_second": 0.481,
       "step": 384
     }
   ],
   "logging_steps": 500,
-  "max_steps": 384,
+  "max_steps": 768,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
-  "total_flos": 942780789120.0,
-  "train_batch_size": 32,
+  "total_flos": 471390394560.0,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9211741662812639,
-    "learning_rate": 0.0002192250105756314,
+    "alpha": 0.9425898264750469,
+    "learning_rate": 0.0005319031539564268,
     "num_train_epochs": 4,
-    "temperature": 2
+    "per_device_train_batch_size": 16,
+    "temperature": 47
   }
 }

run-33/checkpoint-384/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f290a192e0ea1abfff73633e45672101eb50bd64dcaf8d02a78a7004edc1f72b
-size 4920
+oid sha256:135283ce353ecaf9ee2a76e6839a0024571f60628de7c65aedd6fffd84eae3f3
+size 4984