Training in progress, epoch 1

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a42c66a4dbc9c15a1e11e4420cc6cee1f17955c58c47c1150747c7eacb505a57
-size 5998
+oid sha256:06eb43ecf2b1e1f811f42d88309fc77aa4842761b58539fa4ee99a38304ac620
+size 11133

logs/events.out.tfevents.1709698165.a2333cf40ebd.21010.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:150ffbafd9280919c96dc90be5eff3fa0d2c5db60965b58675d31bf084d51b28
+size 5315

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb4d5e48b7b7dcbc4c6a78ba6f64447d41c31642b436b85f9712ebc3f1eb4e10
+oid sha256:a544be95f6f04106deb7c2137fc736ef24e79997e59480896b72c927cd7c7aee
 size 17549312

run-3/checkpoint-384/config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 128,
+  "id2label": {
+    "0": "negative",
+    "1": "positive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 512,
+  "label2id": {
+    "negative": "0",
+    "positive": "1"
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-3/checkpoint-384/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc097722feb26fcf6693b9dbe97ea3a76b6802da9cd21cb5bf38424a0f3cfbad
+size 17549312

run-3/checkpoint-384/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12d4c8e2b1db72df2a15dfb1bed5f04a8214b8e53b3366ebb528306bcc4f8073
+size 35122746

run-3/checkpoint-384/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45b1fc07ed8b121716021ec87c686f1cd9c3b89c82ea08a6a0792d47a39077c9
+size 14054

run-3/checkpoint-384/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9ff4fe586e7eb333ddca8c1956ec7e088a7d846492f389a13a5f39b9e31e62d
+size 1064

run-3/checkpoint-384/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-3/checkpoint-384/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-3/checkpoint-384/trainer_state.json

@@ -0,0 +1,102 @@
+{
+  "best_metric": 0.5009784735812133,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-384",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 384,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.101290702819824,
+      "learning_rate": 1.1962059637335337e-06,
+      "loss": 0.5173,
+      "step": 96
+    },
+    {
+      "epoch": 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
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6350829005241394,
+      "learning_rate": 1.0466802182668422e-06,
+      "loss": 0.4989,
+      "step": 192
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4843444227005871,
+      "eval_f1": 0.6216798277099785,
+      "eval_loss": 0.48798972368240356,
+      "eval_precision": 0.4909297052154195,
+      "eval_recall": 0.8473581213307241,
+      "eval_runtime": 28.492,
+      "eval_samples_per_second": 35.87,
+      "eval_steps_per_second": 1.123,
+      "step": 192
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.8015046119689941,
+      "learning_rate": 8.971544728001503e-07,
+      "loss": 0.4869,
+      "step": 288
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.48238747553816047,
+      "eval_f1": 0.09572649572649572,
+      "eval_loss": 0.4779175817966461,
+      "eval_precision": 0.3783783783783784,
+      "eval_recall": 0.0547945205479452,
+      "eval_runtime": 28.2537,
+      "eval_samples_per_second": 36.172,
+      "eval_steps_per_second": 1.133,
+      "step": 288
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.0801610946655273,
+      "learning_rate": 7.476287273334586e-07,
+      "loss": 0.4795,
+      "step": 384
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5009784735812133,
+      "eval_f1": 0.00390625,
+      "eval_loss": 0.47173693776130676,
+      "eval_precision": 1.0,
+      "eval_recall": 0.0019569471624266144,
+      "eval_runtime": 28.2359,
+      "eval_samples_per_second": 36.195,
+      "eval_steps_per_second": 1.133,
+      "step": 384
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 864,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9,
+  "save_steps": 500,
+  "total_flos": 942780789120.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": {
+    "alpha": 0.586704262013931,
+    "learning_rate": 1.3457317092002255e-06,
+    "num_train_epochs": 9,
+    "temperature": 21
+  }
+}

run-3/checkpoint-384/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
+size 4920

run-3/checkpoint-864/config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 128,
+  "id2label": {
+    "0": "negative",
+    "1": "positive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 512,
+  "label2id": {
+    "negative": "0",
+    "positive": "1"
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-3/checkpoint-864/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80bb08e752a1c5ae2c8f87d95958b8969156c2dbb0cb6ac12ba4be0cee93e0f5
+size 17549312

run-3/checkpoint-864/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95378970087bdbd498cc48c958d686b343b6e916d9c34ba21cfb0ee3000e5f49
+size 35122746

run-3/checkpoint-864/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97d9a51ee84ffde410aa6ee4faf7cd8373bce5e285d974fd9648694f2208055b
+size 14054

run-3/checkpoint-864/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77ee581e56d0666bd21cd21eb09b47ec2b4dafb5bc731d69dde3f5f432c76c82
+size 1064

run-3/checkpoint-864/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-3/checkpoint-864/tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-3/checkpoint-864/trainer_state.json

@@ -0,0 +1,197 @@
+{
+  "best_metric": 0.5009784735812133,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-384",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 864,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.101290702819824,
+      "learning_rate": 1.1962059637335337e-06,
+      "loss": 0.5173,
+      "step": 96
+    },
+    {
+      "epoch": 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
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6350829005241394,
+      "learning_rate": 1.0466802182668422e-06,
+      "loss": 0.4989,
+      "step": 192
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4843444227005871,
+      "eval_f1": 0.6216798277099785,
+      "eval_loss": 0.48798972368240356,
+      "eval_precision": 0.4909297052154195,
+      "eval_recall": 0.8473581213307241,
+      "eval_runtime": 28.492,
+      "eval_samples_per_second": 35.87,
+      "eval_steps_per_second": 1.123,
+      "step": 192
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.8015046119689941,
+      "learning_rate": 8.971544728001503e-07,
+      "loss": 0.4869,
+      "step": 288
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.48238747553816047,
+      "eval_f1": 0.09572649572649572,
+      "eval_loss": 0.4779175817966461,
+      "eval_precision": 0.3783783783783784,
+      "eval_recall": 0.0547945205479452,
+      "eval_runtime": 28.2537,
+      "eval_samples_per_second": 36.172,
+      "eval_steps_per_second": 1.133,
+      "step": 288
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.0801610946655273,
+      "learning_rate": 7.476287273334586e-07,
+      "loss": 0.4795,
+      "step": 384
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5009784735812133,
+      "eval_f1": 0.00390625,
+      "eval_loss": 0.47173693776130676,
+      "eval_precision": 1.0,
+      "eval_recall": 0.0019569471624266144,
+      "eval_runtime": 28.2359,
+      "eval_samples_per_second": 36.195,
+      "eval_steps_per_second": 1.133,
+      "step": 384
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.289751410484314,
+      "learning_rate": 5.981029818667669e-07,
+      "loss": 0.4737,
+      "step": 480
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5009784735812133,
+      "eval_f1": 0.00390625,
+      "eval_loss": 0.468070924282074,
+      "eval_precision": 1.0,
+      "eval_recall": 0.0019569471624266144,
+      "eval_runtime": 28.7218,
+      "eval_samples_per_second": 35.583,
+      "eval_steps_per_second": 1.114,
+      "step": 480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.6734181642532349,
+      "learning_rate": 4.4857723640007514e-07,
+      "loss": 0.4709,
+      "step": 576
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.4659123420715332,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.369,
+      "eval_samples_per_second": 36.025,
+      "eval_steps_per_second": 1.128,
+      "step": 576
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.6338440179824829,
+      "learning_rate": 2.9905149093338343e-07,
+      "loss": 0.4686,
+      "step": 672
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.46463751792907715,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.015,
+      "eval_samples_per_second": 36.48,
+      "eval_steps_per_second": 1.142,
+      "step": 672
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.5559163093566895,
+      "learning_rate": 1.4952574546669171e-07,
+      "loss": 0.4678,
+      "step": 768
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.4639304578304291,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.4868,
+      "eval_samples_per_second": 35.876,
+      "eval_steps_per_second": 1.123,
+      "step": 768
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.9788464307785034,
+      "learning_rate": 0.0,
+      "loss": 0.467,
+      "step": 864
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.4637199938297272,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 31.2173,
+      "eval_samples_per_second": 32.738,
+      "eval_steps_per_second": 1.025,
+      "step": 864
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 864,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9,
+  "save_steps": 500,
+  "total_flos": 2121256775520.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": {
+    "alpha": 0.586704262013931,
+    "learning_rate": 1.3457317092002255e-06,
+    "num_train_epochs": 9,
+    "temperature": 21
+  }
+}

run-3/checkpoint-864/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
+size 4920

run-4/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fab9af68db1fc248c957c6314a56eb88742d6ff2894047fd2dcbac8a3729a080
+oid sha256:a544be95f6f04106deb7c2137fc736ef24e79997e59480896b72c927cd7c7aee
 size 17549312

run-4/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e005ac221b73d6191a7bacfbe571f80dc1eda65b3588beb48f5a09da2bf5a9bc
+oid sha256:92e3700eb6e67af73a6d17192d6328f2e3438132782e383d04facbbbc511695e
 size 35122746

run-4/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6989de7f95b9b4ecddf5ef84ecba9cabb42d72eaa3fb36ceca102c33504152b
+oid sha256:98307cb8361e610ce6078aa8e9ec07ca2d0f6a51244074e95f21bfc44ecff6e0
 size 1064

run-4/checkpoint-96/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.541095890410959,
+  "best_metric": 0.5,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-96",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.20222608745098114,
-      "learning_rate": 0.00010108430383313451,
-      "loss": 0.1454,
+      "grad_norm": 0.6504594683647156,
+      "learning_rate": 7.787665972957169e-06,
+      "loss": 0.59,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.541095890410959,
-      "eval_f1": 0.20642978003384094,
-      "eval_loss": 0.11932022124528885,
-      "eval_precision": 0.7625,
-      "eval_recall": 0.11937377690802348,
-      "eval_runtime": 30.5181,
-      "eval_samples_per_second": 33.488,
-      "eval_steps_per_second": 1.049,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.5820541977882385,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.7531,
+      "eval_samples_per_second": 35.544,
+      "eval_steps_per_second": 1.113,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 672,
+  "max_steps": 192,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.14091216322626776,
-    "learning_rate": 0.00011793168780532361,
-    "num_train_epochs": 7,
-    "temperature": 26
+    "alpha": 0.7979164056999755,
+    "learning_rate": 1.5575331945914338e-05,
+    "num_train_epochs": 2,
+    "temperature": 5
   }
 }

run-4/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e5da8d22bce5e5308ca98e3b419c595cbfb0c0426496b64e62182c8afbfef16
+oid sha256:d75bccfe6c5373ca9b1b18dfaff7dd2ce79806a7acbd4f6cd52c93fdf2f1e00c
 size 4920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
+oid sha256:d75bccfe6c5373ca9b1b18dfaff7dd2ce79806a7acbd4f6cd52c93fdf2f1e00c
 size 4920