jialicheng commited on Nov 12, 2025

Commit

66ff99e

verified ·

1 Parent(s): 1c99a3e

Upload folder using huggingface_hub

Browse files

Files changed (38) hide show

all_results.json +41 -0
mnli-m_results.json +8 -0
mnli-mm_results.json +8 -0
ood_results.json +11 -0
pred_mnli-m.npz.npy +3 -0
pred_mnli-mm.npz.npy +3 -0
pred_ood.npz.npy +3 -0
pred_stress-antonym.npz.npy +3 -0
pred_stress-length_mismatch.npz.npy +3 -0
pred_stress-negation.npz.npy +3 -0
pred_stress-numerical_reasoning.npz.npy +3 -0
pred_stress-spelling_error.npz.npy +3 -0
pred_stress-word_overlap.npz.npy +3 -0
pred_transfer-qnli.npz.npy +3 -0
pred_transfer-snli.npz.npy +3 -0
pred_transfer.npz.npy +3 -0
pytorch_model.bin +3 -0
rep_mnli-m.pt +3 -0
rep_mnli-mm.pt +3 -0
rep_ood.pt +3 -0
rep_transfer.pt +3 -0
special_tokens_map.json +7 -0
stress-antonym_results.json +9 -0
stress-length_mismatch_results.json +9 -0
stress-negation_results.json +9 -0
stress-numerical_reasoning_results.json +9 -0
stress-spelling_error_results.json +9 -0
stress-word_overlap_results.json +9 -0
test_results.json +17 -0
tokenizer.json +0 -0
tokenizer_config.json +16 -0
train_results.json +7 -0
trainer_state.json +493 -0
training_args.bin +3 -0
transfer-qnli_results.json +8 -0
transfer-snli_results.json +8 -0
transfer_results.json +8 -0
vocab.txt +0 -0

all_results.json ADDED Viewed

	@@ -0,0 +1,41 @@

+{
+    "ave/id": 0.7724311220143858,
+    "ave/ood": 0.5464333333333333,
+    "ave/stress": 55.193949185442534,
+    "ave/transfer": 0.7198740093476935,
+    "epoch": 3.0,
+    "eval_accuracy": 0.7198740093476935,
+    "eval_combined_score": 0.7198740093476935,
+    "eval_constituent_accuracy": 0.553,
+    "eval_lexical_overlap_accuracy": 0.5499,
+    "eval_loss": 0.7049594521522522,
+    "eval_runtime": 28.0137,
+    "eval_samples_per_second": 351.328,
+    "eval_steps_per_second": 1.392,
+    "eval_subsequence_accuracy": 0.5364,
+    "gap/constituent": 0.21943112201438575,
+    "gap/lexical_overlap": 0.22253112201438574,
+    "gap/ood": 0.22599778868105247,
+    "gap/stress": 22.049163015996044,
+    "gap/subsequence": 0.2360311220143858,
+    "gap/transfer": 0.052557112666692274,
+    "mnli-m": 0.7697401935812532,
+    "mnli-mm": 0.7751220504475184,
+    "ood": 0.5464333333333333,
+    "ood/constituent": 0.553,
+    "ood/lexical_overlap": 0.5499,
+    "ood/subsequence": 0.5364,
+    "stress-antonym": 60.47405509288918,
+    "stress-length_mismatch": 73.61181864493122,
+    "stress-negation": 49.81692432872254,
+    "stress-numerical_reasoning": 23.235913638757243,
+    "stress-spelling_error": 70.81098828937287,
+    "stress-word_overlap": 53.21399511798211,
+    "train_loss": 1.3483114147227726,
+    "train_runtime": 8047.4713,
+    "train_samples_per_second": 146.395,
+    "train_steps_per_second": 4.575,
+    "transfer": 0.7198740093476935,
+    "transfer-qnli": 50.53999633900788,
+    "transfer-snli": 73.19650477545214
+}

mnli-m_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "eval_accuracy": 0.7697401935812532,
+    "eval_combined_score": 0.7697401935812532,
+    "eval_loss": 0.5902359485626221,
+    "eval_runtime": 28.824,
+    "eval_samples_per_second": 340.515,
+    "eval_steps_per_second": 1.353
+}

mnli-mm_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "eval_accuracy": 0.7751220504475184,
+    "eval_combined_score": 0.7751220504475184,
+    "eval_loss": 0.5796608328819275,
+    "eval_runtime": 28.117,
+    "eval_samples_per_second": 349.682,
+    "eval_steps_per_second": 1.387
+}

ood_results.json ADDED Viewed

	@@ -0,0 +1,11 @@

+{
+    "eval_accuracy": 0.5464333333333333,
+    "eval_combined_score": 0.5464333333333333,
+    "eval_constituent_accuracy": 0.553,
+    "eval_lexical_overlap_accuracy": 0.5499,
+    "eval_loss": 1.4947850704193115,
+    "eval_runtime": 82.1703,
+    "eval_samples_per_second": 365.095,
+    "eval_steps_per_second": 1.436,
+    "eval_subsequence_accuracy": 0.5364
+}

pred_mnli-m.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b2c827d40aff855ced7206c45d6603f017b8083e6b8a54799bf32ff6808fff24
+size 117908

pred_mnli-mm.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f2c2948bba1b39c7178fc327da0493d87c5a21ff95943825fb62ac0b0f08cb80
+size 118112

pred_ood.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ace81bab3dac90958d4772a4e815f2c772e1ecbbe7bac19d9eb32a0ce570fa05
+size 360128

pred_stress-antonym.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f8f580aa6440f7161afbd2997732fa65d1db0c735df2b1d21330dc32c0edbc35
+size 18860

pred_stress-length_mismatch.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ec15f09fff64bc358af1e1fa9b395c6f8b6ada678f616d56dc0d85bb8a262d06
+size 117908

pred_stress-negation.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:074d70d6fa28510795ec8b06629ff6fc422fe428bd4b36d75b73b913859dc22f
+size 118112

pred_stress-numerical_reasoning.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:689b81b2dd7f87c884222aaa7ef86166d5483e4d57dc16d6815f9eac376424eb
+size 91280

pred_stress-spelling_error.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ad5b722d1a43a4bd60fe4930211eb9b7e503c43e78c84c1911efd5e66c2a76b8
+size 109772

pred_stress-word_overlap.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a959b5f8211c0164661973c33d9c43da9ce82510f0127692d1908c663c4be97f
+size 118112

pred_transfer-qnli.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9502b0eb6d0af41c4477325c38489f5e7078f6f7016241907a3afc42b82f4637
+size 65684

pred_transfer-snli.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:fe52c64d1347de6670974bbba0d6c6280be85484da39d58be611b43f852247d2
+size 118232

pred_transfer.npz.npy ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:d41002471efdd62b42f28b85fca8dccac874028e176925a585a6b513f5a6f1bd
+size 118232

pytorch_model.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4045e454d5b9e71f420a70b399a66d4c7dcd60dff51a572119271cb58743b6f8
+size 183452421

rep_mnli-m.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e6859e34497218020e20d18da8466fb198308dcc38cd14229d61b1c2296fcd33
+size 160895222

rep_mnli-mm.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:fa7cb76ed3170e4d31be388f0c6edbc6d4f942a6da1257f023f3d90e98d2d37c
+size 161174066

rep_ood.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b7ad5562a3c4e35e1a5c6a7e3faa2cdd956333647eaa096b8f38c0c4e4e08d8a
+size 491776658

rep_transfer.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:98a689430fe0472bba34abae823a4a0be5d9dea7ff83ee43071ca867245c478a
+size 161338222

special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,7 @@

+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

stress-antonym_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 60.47405509288918,
+    "eval_combined_score": 60.47405509288918,
+    "eval_loss": 1.128919243812561,
+    "eval_runtime": 4.1838,
+    "eval_samples_per_second": 373.109,
+    "eval_steps_per_second": 1.673
+}

stress-length_mismatch_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 73.61181864493122,
+    "eval_combined_score": 73.61181864493122,
+    "eval_loss": 0.6949003338813782,
+    "eval_runtime": 26.2647,
+    "eval_samples_per_second": 373.696,
+    "eval_steps_per_second": 1.485
+}

stress-negation_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 49.81692432872254,
+    "eval_combined_score": 49.81692432872254,
+    "eval_loss": 1.4765657186508179,
+    "eval_runtime": 26.4214,
+    "eval_samples_per_second": 372.122,
+    "eval_steps_per_second": 1.476
+}

stress-numerical_reasoning_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 23.235913638757243,
+    "eval_combined_score": 23.235913638757243,
+    "eval_loss": 1.986377239227295,
+    "eval_runtime": 20.4758,
+    "eval_samples_per_second": 370.975,
+    "eval_steps_per_second": 1.465
+}

stress-spelling_error_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 70.81098828937287,
+    "eval_combined_score": 70.81098828937287,
+    "eval_loss": 0.7184051275253296,
+    "eval_runtime": 24.5771,
+    "eval_samples_per_second": 371.769,
+    "eval_steps_per_second": 1.465
+}

stress-word_overlap_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 3.0,
+    "eval_accuracy": 53.21399511798211,
+    "eval_combined_score": 53.21399511798211,
+    "eval_loss": 1.2522212266921997,
+    "eval_runtime": 26.2796,
+    "eval_samples_per_second": 374.131,
+    "eval_steps_per_second": 1.484
+}

test_results.json ADDED Viewed

	@@ -0,0 +1,17 @@

+{
+    "ave/id": 0.7724311220143858,
+    "ave/ood": 0.5464333333333333,
+    "ave/transfer": 0.7198740093476935,
+    "gap/constituent": 0.21943112201438575,
+    "gap/lexical_overlap": 0.22253112201438574,
+    "gap/ood": 0.22599778868105247,
+    "gap/subsequence": 0.2360311220143858,
+    "gap/transfer": 0.052557112666692274,
+    "mnli-m": 0.7697401935812532,
+    "mnli-mm": 0.7751220504475184,
+    "ood": 0.5464333333333333,
+    "ood/constituent": 0.553,
+    "ood/lexical_overlap": 0.5499,
+    "ood/subsequence": 0.5364,
+    "transfer": 0.7198740093476935
+}

tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "google/bert_uncased_L-8_H-512_A-8",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json ADDED Viewed

	@@ -0,0 +1,7 @@

+{
+    "epoch": 3.0,
+    "train_loss": 1.3483114147227726,
+    "train_runtime": 8047.4713,
+    "train_samples_per_second": 146.395,
+    "train_steps_per_second": 4.575
+}

trainer_state.json ADDED Viewed

	@@ -0,0 +1,493 @@

+{
+  "best_metric": 76.97401935812532,
+  "best_model_checkpoint": "outputs/bert-medium/read/mnli_21/checkpoint-36816",
+  "epoch": 3.0,
+  "global_step": 36816,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9728378965667103e-05,
+      "loss": 1.9992,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 1.8215,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 1.7668,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 1.7235,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 1.6857,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 1.6413,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 1.6325,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 1.6018,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 1.577,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 1.5574,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 1.547,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.5374,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.5273,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.5003,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.4773,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.5036,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.4803,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.4645,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.4562,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.4418,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.438,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.4387,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.4199,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.4249,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 75.63932755985736,
+      "eval_combined_score": 75.63932755985736,
+      "eval_loss": 0.6033145785331726,
+      "eval_runtime": 26.2089,
+      "eval_samples_per_second": 374.491,
+      "eval_steps_per_second": 1.488,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 1.3535,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 1.3334,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 1.3412,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 1.3377,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 1.3265,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 1.3263,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 1.3113,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 1.2968,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 1.3078,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 1.3145,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 1.2913,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 1.2986,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 1.2998,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 1.2958,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 1.3118,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 1.2952,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 1.2899,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 1.2721,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 1.2999,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 1.2829,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 1.2901,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 1.2805,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 1.2658,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 1.2603,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 1.2876,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 75.95517065715741,
+      "eval_combined_score": 75.95517065715741,
+      "eval_loss": 0.5978761315345764,
+      "eval_runtime": 26.1426,
+      "eval_samples_per_second": 375.44,
+      "eval_steps_per_second": 1.492,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 1.2074,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 1.1962,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 1.171,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 1.1868,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 1.1879,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 1.1894,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 1.1795,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 1.164,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 1.1693,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.1696,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 1.1785,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 1.1927,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 1.1839,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 1.1701,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 1.1796,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 1.1791,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 1.1553,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 1.1879,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 1.1746,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 1.1729,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 1.1653,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 1.1793,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 1.1689,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 1.1897,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 76.97401935812532,
+      "eval_combined_score": 76.97401935812532,
+      "eval_loss": 0.5902359485626221,
+      "eval_runtime": 26.2664,
+      "eval_samples_per_second": 373.671,
+      "eval_steps_per_second": 1.485,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 1.3483114147227726,
+      "train_runtime": 8047.4713,
+      "train_samples_per_second": 146.395,
+      "train_steps_per_second": 4.575
+    }
+  ],
+  "max_steps": 36816,
+  "num_train_epochs": 3,
+  "total_flos": 0.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:7a6faa952b42d912dd6f464b441ccb7b1f2a415c7813093150aa93d2070c4d29
+size 4091

transfer-qnli_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "eval_accuracy": 50.53999633900788,
+    "eval_combined_score": 50.53999633900788,
+    "eval_loss": 2.969465494155884,
+    "eval_runtime": 14.5146,
+    "eval_samples_per_second": 376.381,
+    "eval_steps_per_second": 1.516
+}

transfer-snli_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "eval_accuracy": 73.19650477545214,
+    "eval_combined_score": 73.19650477545214,
+    "eval_loss": 0.6977843642234802,
+    "eval_runtime": 27.4004,
+    "eval_samples_per_second": 359.192,
+    "eval_steps_per_second": 1.423
+}

transfer_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "eval_accuracy": 0.7198740093476935,
+    "eval_combined_score": 0.7198740093476935,
+    "eval_loss": 0.7049594521522522,
+    "eval_runtime": 28.0137,
+    "eval_samples_per_second": 351.328,
+    "eval_steps_per_second": 1.392
+}

vocab.txt ADDED Viewed

The diff for this file is too large to render. See raw diff