Training in progress, epoch 1

logs/events.out.tfevents.1709712823.a2333cf40ebd.21010.53

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d953dd5693ad21811d1fd196355dd096862dc8c5967e02a2ec9339794caa6d0
-size 7364
+oid sha256:fb757053de4037fc3be80a48af0f8737bd317f267061b8a20489b6dc43743167
+size 11816

logs/events.out.tfevents.1709713890.a2333cf40ebd.21010.54

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f413e340363ccbb4d232104b0be8e32be3707984308ab944821f7ae35bdb3c58
+size 5314

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6cf86847ee3653d2158930c8c603118e41dea6a5801c888adc67e2fff254632
+oid sha256:e23de7ecf7c1b37935daaff2929e63ea17c1b6baf092aede9d47e318a3d62bfa
 size 17549312

run-0/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-0/checkpoint-864/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a017fbf4b06ddc7a9efc7e8171cad02e61b89fff825afde2c75329d2f3158cb
+size 17549312

run-0/checkpoint-864/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fa11cffc1e42793fa83bb6ac959e629a98060c85610f89f3b595909b1628a27
+size 35122746

run-0/checkpoint-864/rng_state.pth

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

run-0/checkpoint-864/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea23b8eaafb30359f23f51b5426a4f55e46a7d1ea6d6e99187d5e2f35c89e630
+size 1064

run-0/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-0/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-0/checkpoint-864/trainer_state.json

@@ -0,0 +1,197 @@
+{
+  "best_metric": 0.6536203522504892,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-864",
+  "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": 0.8228768110275269,
+      "learning_rate": 1.922542063523116e-05,
+      "loss": 0.7149,
+      "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,
+      "step": 96
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.1306586265563965,
+      "learning_rate": 1.7089262786872146e-05,
+      "loss": 0.7062,
+      "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,
+      "step": 192
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.5551536083221436,
+      "learning_rate": 1.4953104938513126e-05,
+      "loss": 0.6972,
+      "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,
+      "step": 288
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.080883741378784,
+      "learning_rate": 1.2816947090154107e-05,
+      "loss": 0.6854,
+      "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,
+      "step": 384
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.0045652389526367,
+      "learning_rate": 1.068078924179509e-05,
+      "loss": 0.6688,
+      "step": 480
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6105675146771037,
+      "eval_f1": 0.49234693877551017,
+      "eval_loss": 0.6664517521858215,
+      "eval_precision": 0.706959706959707,
+      "eval_recall": 0.3776908023483366,
+      "eval_runtime": 26.8808,
+      "eval_samples_per_second": 38.02,
+      "eval_steps_per_second": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.2470343112945557,
+      "learning_rate": 8.544631393436073e-06,
+      "loss": 0.6549,
+      "step": 576
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6418786692759295,
+      "eval_f1": 0.5906040268456376,
+      "eval_loss": 0.6583118438720703,
+      "eval_precision": 0.6892950391644909,
+      "eval_recall": 0.5166340508806262,
+      "eval_runtime": 26.2817,
+      "eval_samples_per_second": 38.886,
+      "eval_steps_per_second": 1.218,
+      "step": 576
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.420502185821533,
+      "learning_rate": 6.4084735450770535e-06,
+      "loss": 0.6502,
+      "step": 672
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6360078277886497,
+      "eval_f1": 0.5782312925170068,
+      "eval_loss": 0.6568923592567444,
+      "eval_precision": 0.6873315363881402,
+      "eval_recall": 0.49902152641878667,
+      "eval_runtime": 26.5443,
+      "eval_samples_per_second": 38.502,
+      "eval_steps_per_second": 1.206,
+      "step": 672
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.7794277667999268,
+      "learning_rate": 4.2723156967180365e-06,
+      "loss": 0.6453,
+      "step": 768
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.649706457925636,
+      "eval_f1": 0.6108695652173913,
+      "eval_loss": 0.6531394720077515,
+      "eval_precision": 0.687041564792176,
+      "eval_recall": 0.5499021526418787,
+      "eval_runtime": 26.8208,
+      "eval_samples_per_second": 38.105,
+      "eval_steps_per_second": 1.193,
+      "step": 768
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.565406560897827,
+      "learning_rate": 2.1361578483590183e-06,
+      "loss": 0.6432,
+      "step": 864
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6536203522504892,
+      "eval_f1": 0.616052060737527,
+      "eval_loss": 0.6520226001739502,
+      "eval_precision": 0.6909975669099757,
+      "eval_recall": 0.5557729941291585,
+      "eval_runtime": 26.3078,
+      "eval_samples_per_second": 38.848,
+      "eval_steps_per_second": 1.216,
+      "step": 864
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 960,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1984401499680.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": {
+    "alpha": 0.8789746577248793,
+    "learning_rate": 2.136157848359018e-05,
+    "num_train_epochs": 10,
+    "temperature": 14
+  }
+}

run-0/checkpoint-864/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf1d86d1ef9c5f73f9fb51b0ea325423f470c0f4f628eb656cb6d38e041dfed8
+size 4920

run-0/checkpoint-960/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba4a7003755b10592915508f55852913e649127e5440c66527724e592cc2c7b9
+oid sha256:476ee8f1f83a2930e10c4fbd1ec962bea66c39ab816fe16405c1eab1f0bdca1b
 size 17549312

run-0/checkpoint-960/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af588371549f4acb73612d2ad9cd55882aa5fd9f7a0e04c941b81fe140e8d522
+oid sha256:3772077f744c6e14da17813cc0b06415b68919b80b205f25aefe844530eaab5b
 size 35122746

run-0/checkpoint-960/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e74635b58993a273a88d047851a4965dd0f1dab13efd68e1f5c9a3b33d5098c1
+oid sha256:46991e9a48fd9f9c71effd0da425199be1ec600bef86bda24b5504c67c161567
 size 1064

run-0/checkpoint-960/tokenizer.json

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

run-0/checkpoint-960/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8228962818003914,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-768",
+  "best_metric": 0.6536203522504892,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-864",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 960,
@@ -10,192 +10,192 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.8004393577575684,
-      "learning_rate": 0.0007697157586205755,
-      "loss": 0.4935,
+      "grad_norm": 0.8228768110275269,
+      "learning_rate": 1.922542063523116e-05,
+      "loss": 0.7149,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7436399217221135,
-      "eval_f1": 0.7689594356261024,
-      "eval_loss": 0.448881596326828,
-      "eval_precision": 0.6998394863563403,
-      "eval_recall": 0.8532289628180039,
-      "eval_runtime": 28.6476,
-      "eval_samples_per_second": 35.675,
-      "eval_steps_per_second": 1.117,
+      "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,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 4.9879961013793945,
-      "learning_rate": 0.0006841917854405115,
-      "loss": 0.4571,
+      "grad_norm": 2.1306586265563965,
+      "learning_rate": 1.7089262786872146e-05,
+      "loss": 0.7062,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7896281800391389,
-      "eval_f1": 0.8057813911472448,
-      "eval_loss": 0.42557916045188904,
-      "eval_precision": 0.7483221476510067,
-      "eval_recall": 0.87279843444227,
-      "eval_runtime": 28.772,
-      "eval_samples_per_second": 35.521,
-      "eval_steps_per_second": 1.112,
+      "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,
       "step": 192
     },
     {
       "epoch": 3.0,
-      "grad_norm": 3.279979705810547,
-      "learning_rate": 0.0005986678122604475,
-      "loss": 0.4337,
+      "grad_norm": 1.5551536083221436,
+      "learning_rate": 1.4953104938513126e-05,
+      "loss": 0.6972,
       "step": 288
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7945205479452054,
-      "eval_f1": 0.8059149722735675,
-      "eval_loss": 0.4328659176826477,
-      "eval_precision": 0.7635726795096323,
-      "eval_recall": 0.8532289628180039,
-      "eval_runtime": 28.2776,
-      "eval_samples_per_second": 36.142,
-      "eval_steps_per_second": 1.132,
+      "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,
       "step": 288
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.231189727783203,
-      "learning_rate": 0.0005131438390803836,
-      "loss": 0.4235,
+      "grad_norm": 2.080883741378784,
+      "learning_rate": 1.2816947090154107e-05,
+      "loss": 0.6854,
       "step": 384
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7690802348336595,
-      "eval_f1": 0.7704280155642023,
-      "eval_loss": 0.45082738995552063,
-      "eval_precision": 0.7659574468085106,
-      "eval_recall": 0.7749510763209393,
-      "eval_runtime": 29.4798,
-      "eval_samples_per_second": 34.668,
-      "eval_steps_per_second": 1.085,
+      "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,
       "step": 384
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.6898621320724487,
-      "learning_rate": 0.00042761986590031967,
-      "loss": 0.4136,
+      "grad_norm": 2.0045652389526367,
+      "learning_rate": 1.068078924179509e-05,
+      "loss": 0.6688,
       "step": 480
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8033268101761253,
-      "eval_f1": 0.814404432132964,
-      "eval_loss": 0.41852009296417236,
-      "eval_precision": 0.7709790209790209,
-      "eval_recall": 0.863013698630137,
-      "eval_runtime": 29.6952,
-      "eval_samples_per_second": 34.416,
-      "eval_steps_per_second": 1.078,
+      "eval_accuracy": 0.6105675146771037,
+      "eval_f1": 0.49234693877551017,
+      "eval_loss": 0.6664517521858215,
+      "eval_precision": 0.706959706959707,
+      "eval_recall": 0.3776908023483366,
+      "eval_runtime": 26.8808,
+      "eval_samples_per_second": 38.02,
+      "eval_steps_per_second": 1.19,
       "step": 480
     },
     {
       "epoch": 6.0,
-      "grad_norm": 1.360769510269165,
-      "learning_rate": 0.00034209589272025573,
-      "loss": 0.405,
+      "grad_norm": 3.2470343112945557,
+      "learning_rate": 8.544631393436073e-06,
+      "loss": 0.6549,
       "step": 576
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8043052837573386,
-      "eval_f1": 0.8204667863554758,
-      "eval_loss": 0.4140284061431885,
-      "eval_precision": 0.75787728026534,
-      "eval_recall": 0.8943248532289628,
-      "eval_runtime": 28.8106,
-      "eval_samples_per_second": 35.473,
-      "eval_steps_per_second": 1.111,
+      "eval_accuracy": 0.6418786692759295,
+      "eval_f1": 0.5906040268456376,
+      "eval_loss": 0.6583118438720703,
+      "eval_precision": 0.6892950391644909,
+      "eval_recall": 0.5166340508806262,
+      "eval_runtime": 26.2817,
+      "eval_samples_per_second": 38.886,
+      "eval_steps_per_second": 1.218,
       "step": 576
     },
     {
       "epoch": 7.0,
-      "grad_norm": 1.9378271102905273,
-      "learning_rate": 0.0002565719195401918,
-      "loss": 0.3961,
+      "grad_norm": 2.420502185821533,
+      "learning_rate": 6.4084735450770535e-06,
+      "loss": 0.6502,
       "step": 672
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.815068493150685,
-      "eval_f1": 0.8301886792452831,
-      "eval_loss": 0.4103885591030121,
-      "eval_precision": 0.7674418604651163,
-      "eval_recall": 0.9041095890410958,
-      "eval_runtime": 29.2823,
-      "eval_samples_per_second": 34.902,
-      "eval_steps_per_second": 1.093,
+      "eval_accuracy": 0.6360078277886497,
+      "eval_f1": 0.5782312925170068,
+      "eval_loss": 0.6568923592567444,
+      "eval_precision": 0.6873315363881402,
+      "eval_recall": 0.49902152641878667,
+      "eval_runtime": 26.5443,
+      "eval_samples_per_second": 38.502,
+      "eval_steps_per_second": 1.206,
       "step": 672
     },
     {
       "epoch": 8.0,
-      "grad_norm": 1.5738866329193115,
-      "learning_rate": 0.00017104794636012787,
-      "loss": 0.3931,
+      "grad_norm": 2.7794277667999268,
+      "learning_rate": 4.2723156967180365e-06,
+      "loss": 0.6453,
       "step": 768
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8228962818003914,
-      "eval_f1": 0.8340971585701191,
-      "eval_loss": 0.39944925904273987,
-      "eval_precision": 0.7844827586206896,
-      "eval_recall": 0.8904109589041096,
-      "eval_runtime": 29.1614,
-      "eval_samples_per_second": 35.046,
-      "eval_steps_per_second": 1.097,
+      "eval_accuracy": 0.649706457925636,
+      "eval_f1": 0.6108695652173913,
+      "eval_loss": 0.6531394720077515,
+      "eval_precision": 0.687041564792176,
+      "eval_recall": 0.5499021526418787,
+      "eval_runtime": 26.8208,
+      "eval_samples_per_second": 38.105,
+      "eval_steps_per_second": 1.193,
       "step": 768
     },
     {
       "epoch": 9.0,
-      "grad_norm": 0.702528178691864,
-      "learning_rate": 8.552397318006393e-05,
-      "loss": 0.3891,
+      "grad_norm": 2.565406560897827,
+      "learning_rate": 2.1361578483590183e-06,
+      "loss": 0.6432,
       "step": 864
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.8180039138943248,
-      "eval_f1": 0.8287292817679558,
-      "eval_loss": 0.39775392413139343,
-      "eval_precision": 0.782608695652174,
-      "eval_recall": 0.8806262230919765,
-      "eval_runtime": 29.1415,
-      "eval_samples_per_second": 35.07,
-      "eval_steps_per_second": 1.098,
+      "eval_accuracy": 0.6536203522504892,
+      "eval_f1": 0.616052060737527,
+      "eval_loss": 0.6520226001739502,
+      "eval_precision": 0.6909975669099757,
+      "eval_recall": 0.5557729941291585,
+      "eval_runtime": 26.3078,
+      "eval_samples_per_second": 38.848,
+      "eval_steps_per_second": 1.216,
       "step": 864
     },
     {
       "epoch": 10.0,
-      "grad_norm": 2.4524667263031006,
+      "grad_norm": 2.5873706340789795,
       "learning_rate": 0.0,
-      "loss": 0.3819,
+      "loss": 0.645,
       "step": 960
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.8209393346379648,
-      "eval_f1": 0.8288119738072964,
-      "eval_loss": 0.3977544605731964,
-      "eval_precision": 0.7939068100358423,
-      "eval_recall": 0.8669275929549902,
-      "eval_runtime": 30.4608,
-      "eval_samples_per_second": 33.551,
-      "eval_steps_per_second": 1.051,
+      "eval_accuracy": 0.6536203522504892,
+      "eval_f1": 0.6168831168831169,
+      "eval_loss": 0.6515934467315674,
+      "eval_precision": 0.6900726392251816,
+      "eval_recall": 0.5577299412915852,
+      "eval_runtime": 26.5855,
+      "eval_samples_per_second": 38.442,
+      "eval_steps_per_second": 1.204,
       "step": 960
     }
   ],
@@ -204,13 +204,13 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 2356951972800.0,
+  "total_flos": 2204890555200.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.735211359802956,
-    "learning_rate": 0.0008552397318006393,
+    "alpha": 0.8789746577248793,
+    "learning_rate": 2.136157848359018e-05,
     "num_train_epochs": 10,
-    "temperature": 25
+    "temperature": 14
   }
 }

run-0/checkpoint-960/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23804eaaa0feed30486319d2982557f65dce708f7d788864c811c20b2e8c33b4
+oid sha256:bf1d86d1ef9c5f73f9fb51b0ea325423f470c0f4f628eb656cb6d38e041dfed8
 size 4920

run-1/checkpoint-384/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58db24d6a1e94d489cf4b73b5c63cab45aa03d29d732a9d54d6f4f46aef7da02
+oid sha256:7189ccff9386ad17f3ff71b43ae8061f4fea6bd66b1763cfed6af3fbd59e1f22
 size 17549312

run-1/checkpoint-384/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4d03bf37ea38e30d96312f5718b7d70295e30f2d32d7091c3106fe205c75b2d
+oid sha256:afb0b996b74d5df5943878af57f76dd98df2dc9b5ccf5bbe1840dd4b0ed56b28
 size 35122746

run-1/checkpoint-384/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d4b63168300ff1c1b6fb1c3579fe55a55c3b0dedd2cdac5f6ca12f671c781e4
+oid sha256:3be5f1501640a7635c7c786c03440cb1f690d52e629e5cdf40292a361f5a1028
 size 1064

run-1/checkpoint-384/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8140900195694716,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-288",
+  "best_metric": 0.5,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-96",
   "epoch": 4.0,
   "eval_steps": 500,
   "global_step": 384,
@@ -10,93 +10,93 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.6617980003356934,
-      "learning_rate": 0.0006635023570917325,
-      "loss": 0.5037,
+      "grad_norm": 0.4667714238166809,
+      "learning_rate": 7.759069356217338e-06,
+      "loss": 0.317,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7661448140900196,
-      "eval_f1": 0.792714657415438,
-      "eval_loss": 0.4430159628391266,
-      "eval_precision": 0.7118380062305296,
-      "eval_recall": 0.8943248532289628,
-      "eval_runtime": 28.6854,
-      "eval_samples_per_second": 35.628,
-      "eval_steps_per_second": 1.116,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.26975956559181213,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.3197,
+      "eval_samples_per_second": 36.088,
+      "eval_steps_per_second": 1.13,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.12320077419281,
-      "learning_rate": 0.0004976267678187994,
-      "loss": 0.4391,
+      "grad_norm": 0.5722590088844299,
+      "learning_rate": 5.1727129041448925e-06,
+      "loss": 0.267,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8082191780821918,
-      "eval_f1": 0.8227848101265822,
-      "eval_loss": 0.41584911942481995,
-      "eval_precision": 0.7647058823529411,
-      "eval_recall": 0.8904109589041096,
-      "eval_runtime": 28.5215,
-      "eval_samples_per_second": 35.833,
-      "eval_steps_per_second": 1.122,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.2646586298942566,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 29.2799,
+      "eval_samples_per_second": 34.905,
+      "eval_steps_per_second": 1.093,
       "step": 192
     },
     {
       "epoch": 3.0,
-      "grad_norm": 2.946756362915039,
-      "learning_rate": 0.00033175117854586627,
-      "loss": 0.41,
+      "grad_norm": 0.38027867674827576,
+      "learning_rate": 2.5863564520724462e-06,
+      "loss": 0.2634,
       "step": 288
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8140900195694716,
-      "eval_f1": 0.8260073260073261,
-      "eval_loss": 0.43016910552978516,
-      "eval_precision": 0.7762478485370051,
-      "eval_recall": 0.8825831702544031,
-      "eval_runtime": 28.1244,
-      "eval_samples_per_second": 36.339,
-      "eval_steps_per_second": 1.138,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.26031577587127686,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.4418,
+      "eval_samples_per_second": 35.933,
+      "eval_steps_per_second": 1.125,
       "step": 288
     },
     {
       "epoch": 4.0,
-      "grad_norm": 3.868119478225708,
-      "learning_rate": 0.00016587558927293314,
-      "loss": 0.4041,
+      "grad_norm": 0.5131752490997314,
+      "learning_rate": 0.0,
+      "loss": 0.2603,
       "step": 384
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8033268101761253,
-      "eval_f1": 0.802747791952895,
-      "eval_loss": 0.4142443835735321,
-      "eval_precision": 0.8051181102362205,
-      "eval_recall": 0.8003913894324853,
-      "eval_runtime": 29.1758,
-      "eval_samples_per_second": 35.029,
-      "eval_steps_per_second": 1.097,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.2587716281414032,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 28.5668,
+      "eval_samples_per_second": 35.776,
+      "eval_steps_per_second": 1.12,
       "step": 384
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
+  "max_steps": 384,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "total_flos": 942780789120.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7818460169984094,
-    "learning_rate": 0.0008293779463646656,
-    "num_train_epochs": 5,
-    "temperature": 3
+    "alpha": 0.2995113868248127,
+    "learning_rate": 1.0345425808289785e-05,
+    "num_train_epochs": 4,
+    "temperature": 18
   }
 }

run-1/checkpoint-384/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ab2cfe4c2ab241510bc23e9ec5727876e620ab477a4e5ad72a937b25e09df23
+oid sha256:3f7a7027861ec8f0a120b010edf2e889059268bbeac3a62f529a79661a3cf64f
 size 4920

run-1/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65f3dae738929185d1e20f41e7bcf6ea1a73d176be72cfab23bb6503fcf370a1
+oid sha256:e23de7ecf7c1b37935daaff2929e63ea17c1b6baf092aede9d47e318a3d62bfa
 size 17549312

run-1/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5326787fcf5c172156d5e04cae26b95cb125ac94f1539f5594be70c38588eb7d
+oid sha256:ab13633439aabc04ec62e3e3563b84a2145f4f943a616a2c0a606feb7dfe8070
 size 35122746

run-1/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:034a02c034574a59465ffaa25a28c8e0454ff7549227e81b8c6665e0b383867a
+oid sha256:988030091aeb04eea6b1d67ba45707f9a5154327d135a0cced4c586e4e056267
 size 1064

run-1/checkpoint-96/tokenizer.json

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

run-1/checkpoint-96/trainer_state.json

@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.4667714238166809,
-      "learning_rate": 7.759069356217338e-06,
-      "loss": 0.317,
+      "grad_norm": 0.5780554413795471,
+      "learning_rate": 4.753898010566846e-05,
+      "loss": 0.4511,
       "step": 96
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.5,
       "eval_f1": 0.0,
-      "eval_loss": 0.26975956559181213,
+      "eval_loss": 0.2628569006919861,
       "eval_precision": 0.0,
       "eval_recall": 0.0,
-      "eval_runtime": 28.3197,
-      "eval_samples_per_second": 36.088,
-      "eval_steps_per_second": 1.13,
+      "eval_runtime": 27.1178,
+      "eval_samples_per_second": 37.687,
+      "eval_steps_per_second": 1.18,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 384,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 235695197280.0,
+  "total_flos": 220489055520.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2995113868248127,
-    "learning_rate": 1.0345425808289785e-05,
-    "num_train_epochs": 4,
+    "alpha": 0.09512919148833521,
+    "learning_rate": 5.348135261887702e-05,
+    "num_train_epochs": 9,
     "temperature": 18
   }
 }

run-1/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f7a7027861ec8f0a120b010edf2e889059268bbeac3a62f529a79661a3cf64f
+oid sha256:9294acd0d1ac9a0620db6e082aaf47884f1952d159c5d9fb3fb281c25ad02a64
 size 4920

run-2/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f72e368a0735e0614c97feb6fba15c5d298f68104e7128cf582de5807e255027
+oid sha256:57ef7ac15d70c06615ec62775bbd9d47a7a307f789675ebd91d014018802f0ac
 size 17549312

run-2/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b5aece18803da9b55ab957b14fbd401ac7b58d03d968f7138245dd9246cea4e
+oid sha256:8ea9c584a6706cb5bee635fc6f64c59550cbdb80935f1c815de62851495dd156
 size 35122746

run-2/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:773b7d3c411900ce077bec534b9aeadeaba63fc284f46150d41278a6fa17bea6
+oid sha256:d5feae9a058dc367ba00ad758a34b0a64ac558c1f8bd5163d9bb2ce9c13aac22
 size 1064

run-2/checkpoint-96/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5,
+  "best_metric": 0.7984344422700587,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-96",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 3.4041388034820557,
-      "learning_rate": 8.656503841433308e-07,
-      "loss": 0.2746,
+      "grad_norm": 3.4323604106903076,
+      "learning_rate": 0.0004667999543114754,
+      "loss": 0.533,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.6666666666666666,
-      "eval_loss": 0.25451573729515076,
-      "eval_precision": 0.5,
-      "eval_recall": 1.0,
-      "eval_runtime": 28.1178,
-      "eval_samples_per_second": 36.347,
-      "eval_steps_per_second": 1.138,
+      "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
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
+  "max_steps": 192,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.04277939142155329,
-    "learning_rate": 1.0820629801791635e-06,
-    "num_train_epochs": 5,
-    "temperature": 6
+    "alpha": 0.88253398201078,
+    "learning_rate": 0.0009335999086229508,
+    "num_train_epochs": 2,
+    "temperature": 19
   }
 }

run-2/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd5c3fbfe040730ec9bc64036d615ae00f3fe41347ad2df02c2ead7e8faec2da
+oid sha256:2659042d49fa36ce1404b1f7057dd4643dc22f653d1fc44d521700862149e4b7
 size 4920

tokenizer.json

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

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:019a0c30743454578a20e039f50246add333dec6abb36273bd6bc2339b797b34
+oid sha256:9294acd0d1ac9a0620db6e082aaf47884f1952d159c5d9fb3fb281c25ad02a64
 size 4920