Training in progress, epoch 3

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a258d96246086939e3a0b55fad5bb11d9e867a6127846d1046764e9644c17596
+oid sha256:396a602b86e9b47336000b3847f184cbdd8cddba61fcf96401eb71ac6d9cb497
 size 267832560

run-2/checkpoint-25257/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "distilbert-base-uncased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.2",
+  "vocab_size": 30522
+}

run-2/checkpoint-25257/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:396a602b86e9b47336000b3847f184cbdd8cddba61fcf96401eb71ac6d9cb497
+size 267832560

run-2/checkpoint-25257/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cb84843fb9b8c0cd7875414873b061e2e44cb9c82b034b4d12b7d534025c3ac
+size 535727290

run-2/checkpoint-25257/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dc39421d19a2101c52f62917539decd7e32a659ae131a19a96b04ed40a90d15
+size 14244

run-2/checkpoint-25257/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779b39413c9277b7d5bf1aeb14d0ef29a2e7472c45956c49dc655cb0a6f69898
+size 1064

run-2/checkpoint-25257/special_tokens_map.json

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

run-2/checkpoint-25257/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "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_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-25257/trainer_state.json

@@ -0,0 +1,403 @@
+{
+  "best_metric": 0.9059633027522935,
+  "best_model_checkpoint": "distilbert-base-uncased-finetuned-sst2/run-2/checkpoint-25257",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 25257,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05938947618482005,
+      "grad_norm": 6.959882736206055,
+      "learning_rate": 2.20446303880587e-06,
+      "loss": 0.5244,
+      "step": 500
+    },
+    {
+      "epoch": 0.1187789523696401,
+      "grad_norm": 27.900440216064453,
+      "learning_rate": 2.1599410240462895e-06,
+      "loss": 0.3407,
+      "step": 1000
+    },
+    {
+      "epoch": 0.17816842855446016,
+      "grad_norm": 16.400848388671875,
+      "learning_rate": 2.115419009286709e-06,
+      "loss": 0.3444,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2375579047392802,
+      "grad_norm": 16.666898727416992,
+      "learning_rate": 2.0708969945271283e-06,
+      "loss": 0.3153,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2969473809241002,
+      "grad_norm": 22.084596633911133,
+      "learning_rate": 2.026374979767548e-06,
+      "loss": 0.3076,
+      "step": 2500
+    },
+    {
+      "epoch": 0.3563368571089203,
+      "grad_norm": 42.26548385620117,
+      "learning_rate": 1.9818529650079676e-06,
+      "loss": 0.3228,
+      "step": 3000
+    },
+    {
+      "epoch": 0.41572633329374037,
+      "grad_norm": 48.90976333618164,
+      "learning_rate": 1.937330950248387e-06,
+      "loss": 0.3099,
+      "step": 3500
+    },
+    {
+      "epoch": 0.4751158094785604,
+      "grad_norm": 23.298206329345703,
+      "learning_rate": 1.8928089354888066e-06,
+      "loss": 0.2982,
+      "step": 4000
+    },
+    {
+      "epoch": 0.5345052856633804,
+      "grad_norm": 31.678531646728516,
+      "learning_rate": 1.848286920729226e-06,
+      "loss": 0.3012,
+      "step": 4500
+    },
+    {
+      "epoch": 0.5938947618482004,
+      "grad_norm": 14.523043632507324,
+      "learning_rate": 1.8037649059696455e-06,
+      "loss": 0.3075,
+      "step": 5000
+    },
+    {
+      "epoch": 0.6532842380330206,
+      "grad_norm": 2.0264182090759277,
+      "learning_rate": 1.7592428912100647e-06,
+      "loss": 0.2902,
+      "step": 5500
+    },
+    {
+      "epoch": 0.7126737142178406,
+      "grad_norm": 1.4845110177993774,
+      "learning_rate": 1.7147208764504845e-06,
+      "loss": 0.2929,
+      "step": 6000
+    },
+    {
+      "epoch": 0.7720631904026607,
+      "grad_norm": 21.62343406677246,
+      "learning_rate": 1.6701988616909037e-06,
+      "loss": 0.3003,
+      "step": 6500
+    },
+    {
+      "epoch": 0.8314526665874807,
+      "grad_norm": 6.294283866882324,
+      "learning_rate": 1.6256768469313231e-06,
+      "loss": 0.2949,
+      "step": 7000
+    },
+    {
+      "epoch": 0.8908421427723008,
+      "grad_norm": 52.94987106323242,
+      "learning_rate": 1.5811548321717428e-06,
+      "loss": 0.2685,
+      "step": 7500
+    },
+    {
+      "epoch": 0.9502316189571208,
+      "grad_norm": 10.51728343963623,
+      "learning_rate": 1.5366328174121622e-06,
+      "loss": 0.299,
+      "step": 8000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8979357798165137,
+      "eval_loss": 0.33407795429229736,
+      "eval_runtime": 1.2282,
+      "eval_samples_per_second": 709.994,
+      "eval_steps_per_second": 44.782,
+      "step": 8419
+    },
+    {
+      "epoch": 1.0096210951419409,
+      "grad_norm": 0.3329325318336487,
+      "learning_rate": 1.4921108026525816e-06,
+      "loss": 0.2627,
+      "step": 8500
+    },
+    {
+      "epoch": 1.0690105713267608,
+      "grad_norm": 2.096174716949463,
+      "learning_rate": 1.4475887878930012e-06,
+      "loss": 0.2634,
+      "step": 9000
+    },
+    {
+      "epoch": 1.128400047511581,
+      "grad_norm": 6.27714729309082,
+      "learning_rate": 1.4030667731334206e-06,
+      "loss": 0.2496,
+      "step": 9500
+    },
+    {
+      "epoch": 1.187789523696401,
+      "grad_norm": 7.023980140686035,
+      "learning_rate": 1.35854475837384e-06,
+      "loss": 0.2545,
+      "step": 10000
+    },
+    {
+      "epoch": 1.247178999881221,
+      "grad_norm": 4.648576259613037,
+      "learning_rate": 1.3140227436142597e-06,
+      "loss": 0.2437,
+      "step": 10500
+    },
+    {
+      "epoch": 1.3065684760660412,
+      "grad_norm": 9.04944133758545,
+      "learning_rate": 1.269500728854679e-06,
+      "loss": 0.2786,
+      "step": 11000
+    },
+    {
+      "epoch": 1.3659579522508611,
+      "grad_norm": 46.005550384521484,
+      "learning_rate": 1.2249787140950985e-06,
+      "loss": 0.2505,
+      "step": 11500
+    },
+    {
+      "epoch": 1.425347428435681,
+      "grad_norm": 9.347773551940918,
+      "learning_rate": 1.1804566993355181e-06,
+      "loss": 0.253,
+      "step": 12000
+    },
+    {
+      "epoch": 1.4847369046205012,
+      "grad_norm": 0.32879364490509033,
+      "learning_rate": 1.1359346845759375e-06,
+      "loss": 0.2645,
+      "step": 12500
+    },
+    {
+      "epoch": 1.5441263808053214,
+      "grad_norm": 5.314579963684082,
+      "learning_rate": 1.0914126698163572e-06,
+      "loss": 0.245,
+      "step": 13000
+    },
+    {
+      "epoch": 1.6035158569901413,
+      "grad_norm": 109.37725067138672,
+      "learning_rate": 1.0468906550567764e-06,
+      "loss": 0.254,
+      "step": 13500
+    },
+    {
+      "epoch": 1.6629053331749613,
+      "grad_norm": 31.469161987304688,
+      "learning_rate": 1.002368640297196e-06,
+      "loss": 0.2301,
+      "step": 14000
+    },
+    {
+      "epoch": 1.7222948093597814,
+      "grad_norm": 70.80547332763672,
+      "learning_rate": 9.578466255376154e-07,
+      "loss": 0.2229,
+      "step": 14500
+    },
+    {
+      "epoch": 1.7816842855446016,
+      "grad_norm": 0.23098599910736084,
+      "learning_rate": 9.133246107780349e-07,
+      "loss": 0.2528,
+      "step": 15000
+    },
+    {
+      "epoch": 1.8410737617294215,
+      "grad_norm": 32.48334503173828,
+      "learning_rate": 8.688025960184545e-07,
+      "loss": 0.2302,
+      "step": 15500
+    },
+    {
+      "epoch": 1.9004632379142414,
+      "grad_norm": 7.071863651275635,
+      "learning_rate": 8.242805812588738e-07,
+      "loss": 0.2388,
+      "step": 16000
+    },
+    {
+      "epoch": 1.9598527140990618,
+      "grad_norm": 24.40359115600586,
+      "learning_rate": 7.797585664992933e-07,
+      "loss": 0.2594,
+      "step": 16500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9013761467889908,
+      "eval_loss": 0.37052449584007263,
+      "eval_runtime": 1.2139,
+      "eval_samples_per_second": 718.356,
+      "eval_steps_per_second": 45.309,
+      "step": 16838
+    },
+    {
+      "epoch": 2.0192421902838817,
+      "grad_norm": 0.2496776133775711,
+      "learning_rate": 7.352365517397128e-07,
+      "loss": 0.2314,
+      "step": 17000
+    },
+    {
+      "epoch": 2.0786316664687017,
+      "grad_norm": 10.829848289489746,
+      "learning_rate": 6.907145369801322e-07,
+      "loss": 0.2058,
+      "step": 17500
+    },
+    {
+      "epoch": 2.1380211426535216,
+      "grad_norm": 1.5813004970550537,
+      "learning_rate": 6.461925222205517e-07,
+      "loss": 0.2284,
+      "step": 18000
+    },
+    {
+      "epoch": 2.197410618838342,
+      "grad_norm": 1.1905676126480103,
+      "learning_rate": 6.016705074609713e-07,
+      "loss": 0.211,
+      "step": 18500
+    },
+    {
+      "epoch": 2.256800095023162,
+      "grad_norm": 10.768214225769043,
+      "learning_rate": 5.571484927013907e-07,
+      "loss": 0.2349,
+      "step": 19000
+    },
+    {
+      "epoch": 2.316189571207982,
+      "grad_norm": 1.3178966045379639,
+      "learning_rate": 5.126264779418102e-07,
+      "loss": 0.204,
+      "step": 19500
+    },
+    {
+      "epoch": 2.375579047392802,
+      "grad_norm": 89.94540405273438,
+      "learning_rate": 4.6810446318222967e-07,
+      "loss": 0.2094,
+      "step": 20000
+    },
+    {
+      "epoch": 2.434968523577622,
+      "grad_norm": 83.14887237548828,
+      "learning_rate": 4.2358244842264914e-07,
+      "loss": 0.2208,
+      "step": 20500
+    },
+    {
+      "epoch": 2.494357999762442,
+      "grad_norm": 1.2761434316635132,
+      "learning_rate": 3.7906043366306855e-07,
+      "loss": 0.2298,
+      "step": 21000
+    },
+    {
+      "epoch": 2.553747475947262,
+      "grad_norm": 27.40999984741211,
+      "learning_rate": 3.3453841890348807e-07,
+      "loss": 0.2302,
+      "step": 21500
+    },
+    {
+      "epoch": 2.6131369521320824,
+      "grad_norm": 0.43853724002838135,
+      "learning_rate": 2.9001640414390754e-07,
+      "loss": 0.2238,
+      "step": 22000
+    },
+    {
+      "epoch": 2.6725264283169023,
+      "grad_norm": 0.7176188826560974,
+      "learning_rate": 2.45494389384327e-07,
+      "loss": 0.2256,
+      "step": 22500
+    },
+    {
+      "epoch": 2.7319159045017223,
+      "grad_norm": 1.2089786529541016,
+      "learning_rate": 2.009723746247465e-07,
+      "loss": 0.2053,
+      "step": 23000
+    },
+    {
+      "epoch": 2.791305380686542,
+      "grad_norm": 48.92116165161133,
+      "learning_rate": 1.5645035986516597e-07,
+      "loss": 0.2301,
+      "step": 23500
+    },
+    {
+      "epoch": 2.850694856871362,
+      "grad_norm": 1.1826889514923096,
+      "learning_rate": 1.1192834510558544e-07,
+      "loss": 0.208,
+      "step": 24000
+    },
+    {
+      "epoch": 2.9100843330561825,
+      "grad_norm": 16.557043075561523,
+      "learning_rate": 6.740633034600492e-08,
+      "loss": 0.2303,
+      "step": 24500
+    },
+    {
+      "epoch": 2.9694738092410025,
+      "grad_norm": 1.251015067100525,
+      "learning_rate": 2.2884315586424388e-08,
+      "loss": 0.2196,
+      "step": 25000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9059633027522935,
+      "eval_loss": 0.37748080492019653,
+      "eval_runtime": 1.2054,
+      "eval_samples_per_second": 723.412,
+      "eval_steps_per_second": 45.628,
+      "step": 25257
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 25257,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1538900160176328.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 2.2489850535654507e-06,
+    "num_train_epochs": 3,
+    "per_device_train_batch_size": 8,
+    "seed": 11
+  }
+}

run-2/checkpoint-25257/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f28c54c631beb4aad19ce06fef9fbf9a411e80177ae39964fa9c31526a231b3
+size 5048

runs/May14_13-42-22_d0b6fd4b0e89/events.out.tfevents.1715698909.d0b6fd4b0e89.396.4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:259da8a5606b186ca87058a59262f9bd59eac44c18b7cd16864fa1d225ae7d6f
-size 12687
+oid sha256:f1aa58c4bedb0196a81dece0046e420a287958d5b1e84e4ab9efb7732834cba5
+size 16601