Training in progress, epoch 1

logs/events.out.tfevents.1709737211.1dcff7455174.7719.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:547e623b37b94297fc4f9d015128dc831f3e7facd17e094ec7db4df88b330c86
-size 5998
+oid sha256:66770da5ecd6e427377d853e73614d249ab034e9e3a9fa0401c8a4433d63a7da
+size 8401

logs/events.out.tfevents.1709737723.1dcff7455174.7719.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdbda6302dedd3c7a125c014539e9be2875b24d549ede70f091bc89bed984ea6
+size 5315

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57c2d3f61a249e6d0cec071823ce3825a3e493a94eb26095a8de3b10c6e69187
+oid sha256:f580c7021c521e50c4f09989078be6521aed1a3504233121cb4128acd7ffaf3f
 size 17549312

run-0/checkpoint-384/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a5cced5d1408a3231d9adec89e92636cb688ecf5bcd1f99c9679e57f7106218
+oid sha256:1bce608ff84d676ff00ffe40bd50b97fbc7541a452729c9f63e6157c1aa613a0
 size 17549312

run-0/checkpoint-384/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0e3db2ea3c5d266e5b6785e41f67a9da15b4493f25e13329b02022502d9e2e7
+oid sha256:39a4a9968d3befda2b3b1fc8eb232f4dd5dd4b1439f48ad4b5bd16b2994c1dd9
 size 35122746

run-0/checkpoint-384/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2efd5220d80201e72fefaf32ad5da346b0aa9bc0a3be9b0c7b351c23430409f
+oid sha256:36f2ca0dfd9a3945022ef05581370b90827ced4c3be77e49482b876673eb94ec
 size 1064

run-0/checkpoint-384/tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 29,
+    "max_length": 31,
     "strategy": "LongestFirst",
     "stride": 0
   },

run-0/checkpoint-384/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5606653620352251,
+  "best_metric": 0.8003913894324853,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-384",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -10,93 +10,93 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.8228768110275269,
-      "learning_rate": 1.922542063523116e-05,
-      "loss": 0.7149,
+      "grad_norm": 1.6740443706512451,
+      "learning_rate": 0.0001218354408608861,
+      "loss": 0.4816,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.007766990291262136,
-      "eval_loss": 0.7131175398826599,
-      "eval_precision": 0.5,
-      "eval_recall": 0.003913894324853229,
-      "eval_runtime": 26.5316,
-      "eval_samples_per_second": 38.52,
-      "eval_steps_per_second": 1.206,
+      "eval_accuracy": 0.7514677103718199,
+      "eval_f1": 0.7519531250000001,
+      "eval_loss": 0.41538161039352417,
+      "eval_precision": 0.7504873294346979,
+      "eval_recall": 0.7534246575342466,
+      "eval_runtime": 26.5175,
+      "eval_samples_per_second": 38.541,
+      "eval_steps_per_second": 1.207,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.1306586265563965,
-      "learning_rate": 1.7089262786872146e-05,
-      "loss": 0.7062,
+      "grad_norm": 1.7866544723510742,
+      "learning_rate": 9.137658064566457e-05,
+      "loss": 0.4182,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5136986301369864,
-      "eval_f1": 0.08133086876155268,
-      "eval_loss": 0.7069234848022461,
-      "eval_precision": 0.7333333333333333,
-      "eval_recall": 0.043052837573385516,
-      "eval_runtime": 26.8847,
-      "eval_samples_per_second": 38.014,
-      "eval_steps_per_second": 1.19,
+      "eval_accuracy": 0.7798434442270059,
+      "eval_f1": 0.7817652764306499,
+      "eval_loss": 0.3980446457862854,
+      "eval_precision": 0.775,
+      "eval_recall": 0.7886497064579256,
+      "eval_runtime": 25.5088,
+      "eval_samples_per_second": 40.065,
+      "eval_steps_per_second": 1.254,
       "step": 192
     },
     {
       "epoch": 3.0,
-      "grad_norm": 1.5551536083221436,
-      "learning_rate": 1.4953104938513126e-05,
-      "loss": 0.6972,
+      "grad_norm": 2.1238555908203125,
+      "learning_rate": 6.091772043044305e-05,
+      "loss": 0.4044,
       "step": 288
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.512720156555773,
-      "eval_f1": 0.08791208791208792,
-      "eval_loss": 0.6998902559280396,
-      "eval_precision": 0.6857142857142857,
-      "eval_recall": 0.046966731898238745,
-      "eval_runtime": 26.6683,
-      "eval_samples_per_second": 38.323,
-      "eval_steps_per_second": 1.2,
+      "eval_accuracy": 0.7984344422700587,
+      "eval_f1": 0.8205574912891985,
+      "eval_loss": 0.39847832918167114,
+      "eval_precision": 0.7394034536891679,
+      "eval_recall": 0.9217221135029354,
+      "eval_runtime": 25.6502,
+      "eval_samples_per_second": 39.844,
+      "eval_steps_per_second": 1.248,
       "step": 288
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.080883741378784,
-      "learning_rate": 1.2816947090154107e-05,
-      "loss": 0.6854,
+      "grad_norm": 2.6613142490386963,
+      "learning_rate": 3.0458860215221525e-05,
+      "loss": 0.3971,
       "step": 384
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5606653620352251,
-      "eval_f1": 0.3145038167938931,
-      "eval_loss": 0.6807624697685242,
-      "eval_precision": 0.7152777777777778,
-      "eval_recall": 0.20156555772994128,
-      "eval_runtime": 26.6572,
-      "eval_samples_per_second": 38.339,
-      "eval_steps_per_second": 1.2,
+      "eval_accuracy": 0.8003913894324853,
+      "eval_f1": 0.799212598425197,
+      "eval_loss": 0.38756656646728516,
+      "eval_precision": 0.803960396039604,
+      "eval_recall": 0.7945205479452054,
+      "eval_runtime": 26.1487,
+      "eval_samples_per_second": 39.084,
+      "eval_steps_per_second": 1.224,
       "step": 384
     }
   ],
   "logging_steps": 500,
-  "max_steps": 960,
+  "max_steps": 480,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 881956222080.0,
+  "total_flos": 942780789120.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8789746577248793,
-    "learning_rate": 2.136157848359018e-05,
-    "num_train_epochs": 10,
-    "temperature": 14
+    "alpha": 0.679174768290245,
+    "learning_rate": 0.00015229430107610762,
+    "num_train_epochs": 5,
+    "temperature": 27
   }
 }

run-0/checkpoint-384/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf1d86d1ef9c5f73f9fb51b0ea325423f470c0f4f628eb656cb6d38e041dfed8
+oid sha256:1873f3a503b2d5b3f7200baa33f737ba864dd65edb3834d85c5b8e40b6b72f07
 size 4920

run-0/checkpoint-480/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-0/checkpoint-480/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef39c1bfae33e269d63eca2dba3acdfd058e775356849d8caa533434a56a2116
+size 17549312

run-0/checkpoint-480/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2773d01ed3883b1e485724aa4513545bf7e3fbdc7a714d33c0bda776eb6e3bd
+size 35122746

run-0/checkpoint-480/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d93f4cfe67413cd27374342b3b9f13806d5fb593a9dc59106a8b24dedb590a8
+size 14054

run-0/checkpoint-480/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7803f9f54b1d30ab07b891855d766e197fb736184fcc7dedb2e51526851226c1
+size 1064

run-0/checkpoint-480/special_tokens_map.json

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

run-0/checkpoint-480/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-0/checkpoint-480/trainer_state.json

@@ -0,0 +1,121 @@
+{
+  "best_metric": 0.8170254403131115,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-480",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 480,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.6740443706512451,
+      "learning_rate": 0.0001218354408608861,
+      "loss": 0.4816,
+      "step": 96
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7514677103718199,
+      "eval_f1": 0.7519531250000001,
+      "eval_loss": 0.41538161039352417,
+      "eval_precision": 0.7504873294346979,
+      "eval_recall": 0.7534246575342466,
+      "eval_runtime": 26.5175,
+      "eval_samples_per_second": 38.541,
+      "eval_steps_per_second": 1.207,
+      "step": 96
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.7866544723510742,
+      "learning_rate": 9.137658064566457e-05,
+      "loss": 0.4182,
+      "step": 192
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7798434442270059,
+      "eval_f1": 0.7817652764306499,
+      "eval_loss": 0.3980446457862854,
+      "eval_precision": 0.775,
+      "eval_recall": 0.7886497064579256,
+      "eval_runtime": 25.5088,
+      "eval_samples_per_second": 40.065,
+      "eval_steps_per_second": 1.254,
+      "step": 192
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.1238555908203125,
+      "learning_rate": 6.091772043044305e-05,
+      "loss": 0.4044,
+      "step": 288
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7984344422700587,
+      "eval_f1": 0.8205574912891985,
+      "eval_loss": 0.39847832918167114,
+      "eval_precision": 0.7394034536891679,
+      "eval_recall": 0.9217221135029354,
+      "eval_runtime": 25.6502,
+      "eval_samples_per_second": 39.844,
+      "eval_steps_per_second": 1.248,
+      "step": 288
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.6613142490386963,
+      "learning_rate": 3.0458860215221525e-05,
+      "loss": 0.3971,
+      "step": 384
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8003913894324853,
+      "eval_f1": 0.799212598425197,
+      "eval_loss": 0.38756656646728516,
+      "eval_precision": 0.803960396039604,
+      "eval_recall": 0.7945205479452054,
+      "eval_runtime": 26.1487,
+      "eval_samples_per_second": 39.084,
+      "eval_steps_per_second": 1.224,
+      "step": 384
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.489380955696106,
+      "learning_rate": 0.0,
+      "loss": 0.3912,
+      "step": 480
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8170254403131115,
+      "eval_f1": 0.8270120259019426,
+      "eval_loss": 0.3846189081668854,
+      "eval_precision": 0.7842105263157895,
+      "eval_recall": 0.8747553816046967,
+      "eval_runtime": 25.7274,
+      "eval_samples_per_second": 39.724,
+      "eval_steps_per_second": 1.244,
+      "step": 480
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 480,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 1178475986400.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": {
+    "alpha": 0.679174768290245,
+    "learning_rate": 0.00015229430107610762,
+    "num_train_epochs": 5,
+    "temperature": 27
+  }
+}

run-0/checkpoint-480/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1873f3a503b2d5b3f7200baa33f737ba864dd65edb3834d85c5b8e40b6b72f07
+size 4920

run-1/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e23de7ecf7c1b37935daaff2929e63ea17c1b6baf092aede9d47e318a3d62bfa
+oid sha256:f580c7021c521e50c4f09989078be6521aed1a3504233121cb4128acd7ffaf3f
 size 17549312

run-1/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab13633439aabc04ec62e3e3563b84a2145f4f943a616a2c0a606feb7dfe8070
+oid sha256:91e888b82282d118e9145b95f3a0386ef58f658c83e89e13b16215c4e2411c83
 size 35122746

run-1/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:988030091aeb04eea6b1d67ba45707f9a5154327d135a0cced4c586e4e056267
+oid sha256:11606a3fe2610be2878eac5b9a478bb3687282988a72c7cf16c2b9960e57533c
 size 1064

run-1/checkpoint-96/tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 29,
+    "max_length": 31,
     "strategy": "LongestFirst",
     "stride": 0
   },

run-1/checkpoint-96/trainer_state.json

@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.5780554413795471,
-      "learning_rate": 4.753898010566846e-05,
-      "loss": 0.4511,
+      "grad_norm": 0.40490686893463135,
+      "learning_rate": 1.4953262133881219e-05,
+      "loss": 0.1703,
       "step": 96
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.5,
       "eval_f1": 0.0,
-      "eval_loss": 0.2628569006919861,
+      "eval_loss": 0.11485283076763153,
       "eval_precision": 0.0,
       "eval_recall": 0.0,
-      "eval_runtime": 27.1178,
-      "eval_samples_per_second": 37.687,
-      "eval_steps_per_second": 1.18,
+      "eval_runtime": 25.7126,
+      "eval_samples_per_second": 39.747,
+      "eval_steps_per_second": 1.245,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 864,
+  "max_steps": 768,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 9,
+  "num_train_epochs": 8,
   "save_steps": 500,
-  "total_flos": 220489055520.0,
+  "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.09512919148833521,
-    "learning_rate": 5.348135261887702e-05,
-    "num_train_epochs": 9,
-    "temperature": 18
+    "alpha": 0.09656764013507046,
+    "learning_rate": 1.7089442438721394e-05,
+    "num_train_epochs": 8,
+    "temperature": 20
   }
 }

run-1/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9294acd0d1ac9a0620db6e082aaf47884f1952d159c5d9fb3fb281c25ad02a64
+oid sha256:40c83b924ef70318ad972d6c481c684f04bf59974ceae562e8816ac4c289de55
 size 4920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c234fefe2007f16535ef723e1425baa7c9b3f0aec93243168de7f22031da2d1
+oid sha256:40c83b924ef70318ad972d6c481c684f04bf59974ceae562e8816ac4c289de55
 size 4920