jialicheng commited on Nov 11, 2025

Commit

214dc80

verified ·

1 Parent(s): 3ed6748

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 +14 -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": 0.8307649976559848,
+    "ave/ood": 0.7316666666666667,
+    "ave/stress": 70.12055737749229,
+    "ave/transfer": 0.7955700060963219,
+    "epoch": 3.0,
+    "eval_accuracy": 0.7955700060963219,
+    "eval_combined_score": 0.7955700060963219,
+    "eval_constituent_accuracy": 0.7032,
+    "eval_lexical_overlap_accuracy": 0.7566,
+    "eval_loss": 0.5698015689849854,
+    "eval_runtime": 243.8952,
+    "eval_samples_per_second": 40.353,
+    "eval_steps_per_second": 0.16,
+    "eval_subsequence_accuracy": 0.7352,
+    "gap/constituent": 0.12756499765598472,
+    "gap/lexical_overlap": 0.07416499765598472,
+    "gap/ood": 0.09909833098931808,
+    "gap/stress": 12.95594238810618,
+    "gap/subsequence": 0.0955649976559848,
+    "gap/transfer": 0.03519499155966288,
+    "mnli-m": 0.833112582781457,
+    "mnli-mm": 0.8284174125305126,
+    "ood": 0.7316666666666667,
+    "ood/constituent": 0.7032,
+    "ood/lexical_overlap": 0.7566,
+    "ood/subsequence": 0.7352,
+    "stress-antonym": 72.13324791800129,
+    "stress-length_mismatch": 81.23280692817116,
+    "stress-negation": 63.883238405207486,
+    "stress-numerical_reasoning": 59.79462875197472,
+    "stress-spelling_error": 81.08788442596038,
+    "stress-word_overlap": 62.59153783563873,
+    "train_loss": 0.9035921859409642,
+    "train_runtime": 54432.0817,
+    "train_samples_per_second": 21.644,
+    "train_steps_per_second": 0.676,
+    "transfer": 0.7955700060963219,
+    "transfer-qnli": 50.33864177192019,
+    "transfer-snli": 81.92440560861614
+}

mnli-m_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.833112582781457,
+    "eval_combined_score": 0.833112582781457,
+    "eval_loss": 0.4648875296115875,
+    "eval_runtime": 243.9382,
+    "eval_samples_per_second": 40.236,
+    "eval_steps_per_second": 0.16
+}

mnli-mm_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.8284174125305126,
+    "eval_combined_score": 0.8284174125305126,
+    "eval_loss": 0.466293603181839,
+    "eval_runtime": 243.6578,
+    "eval_samples_per_second": 40.352,
+    "eval_steps_per_second": 0.16
+}

ood_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.7316666666666667,
+    "eval_combined_score": 0.7316666666666667,
+    "eval_constituent_accuracy": 0.7032,
+    "eval_lexical_overlap_accuracy": 0.7566,
+    "eval_loss": 1.0719795227050781,
+    "eval_runtime": 738.574,
+    "eval_samples_per_second": 40.619,
+    "eval_steps_per_second": 0.16,
+    "eval_subsequence_accuracy": 0.7352
+}

pred_mnli-m.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:7725ea53560551f5882a739c12c209940868300881a53e0d8c45f31dc6509cd7
+size 117908

pred_mnli-mm.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:e904c3995d5d9e7b459f64d5067dcf175234ae9ad5469991f000da09d527ea7d
+size 118112

pred_ood.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:677dc5d4bfa4fa2006018ffa5071d8a58391a55e441710d4cf7e429a03eca2f2
+size 360128

pred_stress-antonym.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:2f73f7f804ee39f53d9c7e5cd957677cfce76c494566f295f07a3488c67b1bed
+size 18860

pred_stress-length_mismatch.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:25a2f0e6ced8a0140085c5372cc612884996173473e415412db1ed464edbde0d
+size 117908

pred_stress-negation.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:e962e945053747753c4c60e15575c08688a59f1730e3e4a692ee22dc67082dbf
+size 118112

pred_stress-numerical_reasoning.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:ac3549da4b11e331039a5b2db6ffeba1914f01b89bb5ad181f483fb02fd0d1ea
+size 91280

pred_stress-spelling_error.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:19eeafd60b8c8b73e7fcc262542600a6d025a11dafc01ee184708aa6320a21f7
+size 109772

pred_stress-word_overlap.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:caa6d0c869dfb4851b609b000002908217c41c02cf4dd67b721814500a801119
+size 118112

pred_transfer-qnli.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:37d6250f62c66ddc893294985feef1802be7129ce0d08bab1cb2d7504faf3dc2
+size 65684

pred_transfer-snli.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:61537e0edbd6033192267caba0c4945dc362f2267d563c8b34464dc2525c31e3
+size 118232

pred_transfer.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:6c44b02cc3768a0bbe428d3b44c09d4882f781a12d0cda6a6efd33f947e521fe
+size 118232

pytorch_model.bin ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a37ea9b0917fbd49101c011d03ba9b2739ef47a31f4da83fb4cfc3280ce85101
+size 1546553029

rep_mnli-m.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:723d726f77651f51e799c2ea34297c2595c07009686e7c4455d7b46734bb3323
+size 965112614

rep_mnli-mm.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:f75066b1f6a170119bae7d3a35152dca8610904a7e40404ccef7fcead5a07336
+size 966784722

rep_ood.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:8376c09a18396a9138fbe6492eb66773ea702aa26798115660cba288f027a9fc
+size 2949894778

rep_transfer.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:dbde7af1bc1e4de42a92e87c27196a513e09b69ddc1ee9aa101a59caad2da2fb
+size 967768702

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": 72.13324791800129,
+    "eval_combined_score": 72.13324791800129,
+    "eval_loss": 0.8699958324432373,
+    "eval_runtime": 19.557,
+    "eval_samples_per_second": 79.818,
+    "eval_steps_per_second": 0.358
+}

stress-length_mismatch_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 81.23280692817116,
+    "eval_combined_score": 81.23280692817116,
+    "eval_loss": 0.5130394101142883,
+    "eval_runtime": 122.9898,
+    "eval_samples_per_second": 79.803,
+    "eval_steps_per_second": 0.317
+}

stress-negation_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 63.883238405207486,
+    "eval_combined_score": 63.883238405207486,
+    "eval_loss": 0.7891832590103149,
+    "eval_runtime": 123.0413,
+    "eval_samples_per_second": 79.908,
+    "eval_steps_per_second": 0.317
+}

stress-numerical_reasoning_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 59.79462875197472,
+    "eval_combined_score": 59.79462875197472,
+    "eval_loss": 0.9163689017295837,
+    "eval_runtime": 95.0023,
+    "eval_samples_per_second": 79.956,
+    "eval_steps_per_second": 0.316
+}

stress-spelling_error_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 81.08788442596038,
+    "eval_combined_score": 81.08788442596038,
+    "eval_loss": 0.5173742771148682,
+    "eval_runtime": 114.117,
+    "eval_samples_per_second": 80.067,
+    "eval_steps_per_second": 0.315
+}

stress-word_overlap_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 62.59153783563873,
+    "eval_combined_score": 62.59153783563873,
+    "eval_loss": 0.8422909379005432,
+    "eval_runtime": 123.2158,
+    "eval_samples_per_second": 79.795,
+    "eval_steps_per_second": 0.317
+}

test_results.json ADDED Viewed

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

+{
+    "ave/id": 0.8307649976559848,
+    "ave/ood": 0.7316666666666667,
+    "ave/transfer": 0.7955700060963219,
+    "gap/constituent": 0.12756499765598472,
+    "gap/lexical_overlap": 0.07416499765598472,
+    "gap/ood": 0.09909833098931808,
+    "gap/subsequence": 0.0955649976559848,
+    "gap/transfer": 0.03519499155966288,
+    "mnli-m": 0.833112582781457,
+    "mnli-mm": 0.8284174125305126,
+    "ood": 0.7316666666666667,
+    "ood/constituent": 0.7032,
+    "ood/lexical_overlap": 0.7566,
+    "ood/subsequence": 0.7352,
+    "transfer": 0.7955700060963219
+}

tokenizer.json ADDED Viewed

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

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,14 @@

+{
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "name_or_path": "bert-large-uncased",
+  "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": 0.9035921859409642,
+    "train_runtime": 54432.0817,
+    "train_samples_per_second": 21.644,
+    "train_steps_per_second": 0.676
+}

trainer_state.json ADDED Viewed

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

+{
+  "best_metric": 83.3112582781457,
+  "best_model_checkpoint": "outputs/bert-large/read/mnli_21/checkpoint-24544",
+  "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": 1.8184,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 1.498,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 1.4285,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 1.3852,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 1.3299,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 1.2948,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 1.2937,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 1.2636,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 1.2317,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 1.2179,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 1.1939,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.1994,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.1763,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.1584,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.1315,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.1664,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.1681,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.1406,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.1249,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.1217,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.1059,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.1097,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.0972,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.1017,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 82.04788588894549,
+      "eval_combined_score": 82.04788588894549,
+      "eval_loss": 0.4863318204879761,
+      "eval_runtime": 123.0921,
+      "eval_samples_per_second": 79.737,
+      "eval_steps_per_second": 0.317,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 0.9725,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 0.8703,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 0.8647,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 0.8809,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 0.8799,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 0.8643,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 0.8589,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 0.8477,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 0.8672,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 0.8524,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 0.8539,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 0.8527,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 0.846,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 0.8653,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 0.8761,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 0.8469,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 0.8452,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 0.8363,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 0.8538,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 0.8525,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 0.8526,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 0.8455,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 0.8346,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 0.8374,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 0.8537,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 83.3112582781457,
+      "eval_combined_score": 83.3112582781457,
+      "eval_loss": 0.4648875296115875,
+      "eval_runtime": 123.0335,
+      "eval_samples_per_second": 79.775,
+      "eval_steps_per_second": 0.317,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 0.6479,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 0.6324,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 0.6072,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 0.6265,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 0.6194,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 0.6223,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 0.63,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 0.621,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 0.6153,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 0.6248,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 0.6229,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 0.6209,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 0.6304,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 0.5968,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 0.6269,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 0.6263,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 0.6111,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 0.6349,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 0.608,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 0.6201,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 0.6071,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 0.6065,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 0.6067,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 0.6216,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 82.79164544065206,
+      "eval_combined_score": 82.79164544065206,
+      "eval_loss": 0.5364694595336914,
+      "eval_runtime": 123.3628,
+      "eval_samples_per_second": 79.562,
+      "eval_steps_per_second": 0.316,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 0.9035921859409642,
+      "train_runtime": 54432.0817,
+      "train_samples_per_second": 21.644,
+      "train_steps_per_second": 0.676
+    }
+  ],
+  "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 sha256:562914c06761b1def459f6246ac3325ba226a21ffde5093b20e833a616c05565
+size 4091

transfer-qnli_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 50.33864177192019,
+    "eval_combined_score": 50.33864177192019,
+    "eval_loss": 4.116509437561035,
+    "eval_runtime": 61.8793,
+    "eval_samples_per_second": 88.285,
+    "eval_steps_per_second": 0.356
+}

transfer-snli_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 81.92440560861614,
+    "eval_combined_score": 81.92440560861614,
+    "eval_loss": 0.7006551027297974,
+    "eval_runtime": 112.1367,
+    "eval_samples_per_second": 87.768,
+    "eval_steps_per_second": 0.348
+}

transfer_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.7955700060963219,
+    "eval_combined_score": 0.7955700060963219,
+    "eval_loss": 0.5698015689849854,
+    "eval_runtime": 243.8952,
+    "eval_samples_per_second": 40.353,
+    "eval_steps_per_second": 0.16
+}

vocab.txt ADDED Viewed

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