jialicheng commited on Nov 11, 2025

Commit

8bc0ed7

verified ·

1 Parent(s): 9572316

Upload folder using huggingface_hub

Browse files

Files changed (39) 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.npz +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": 0.6483877278886412,
+    "ave/ood": 0.5,
+    "ave/stress": 42.02114161047434,
+    "ave/transfer": 0.5376955903271693,
+    "epoch": 3.0,
+    "eval_accuracy": 0.5376955903271693,
+    "eval_combined_score": 0.5376955903271693,
+    "eval_constituent_accuracy": 0.5,
+    "eval_lexical_overlap_accuracy": 0.5,
+    "eval_loss": 0.8987720608711243,
+    "eval_runtime": 3.7795,
+    "eval_samples_per_second": 2604.049,
+    "eval_steps_per_second": 10.319,
+    "eval_subsequence_accuracy": 0.5,
+    "gap/constituent": 0.14838772788864119,
+    "gap/lexical_overlap": 0.14838772788864119,
+    "gap/ood": 0.14838772788864119,
+    "gap/stress": 22.817631178389796,
+    "gap/subsequence": 0.14838772788864119,
+    "gap/transfer": 0.11069213756147189,
+    "mnli-m": 0.6409577177789099,
+    "mnli-mm": 0.6558177379983726,
+    "ood": 0.5,
+    "ood/constituent": 0.5,
+    "ood/lexical_overlap": 0.5,
+    "ood/subsequence": 0.5,
+    "stress-antonym": 4.099935938500961,
+    "stress-length_mismatch": 60.69281711665817,
+    "stress-negation": 43.480471928397066,
+    "stress-numerical_reasoning": 28.001579778830965,
+    "stress-spelling_error": 59.363029440735474,
+    "stress-word_overlap": 56.489015459723355,
+    "train_loss": 1.8895709598546233,
+    "train_runtime": 3349.1082,
+    "train_samples_per_second": 351.767,
+    "train_steps_per_second": 10.993,
+    "transfer": 0.5376955903271693,
+    "transfer-qnli": 51.601684056379284,
+    "transfer-snli": 53.73907742328795
+}

mnli-m_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.6409577177789099,
+    "eval_combined_score": 0.6409577177789099,
+    "eval_loss": 0.8214038610458374,
+    "eval_runtime": 5.0137,
+    "eval_samples_per_second": 1957.649,
+    "eval_steps_per_second": 7.779
+}

mnli-mm_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.6558177379983726,
+    "eval_combined_score": 0.6558177379983726,
+    "eval_loss": 0.8016026020050049,
+    "eval_runtime": 4.0263,
+    "eval_samples_per_second": 2441.921,
+    "eval_steps_per_second": 9.686
+}

ood_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.5,
+    "eval_combined_score": 0.5,
+    "eval_constituent_accuracy": 0.5,
+    "eval_lexical_overlap_accuracy": 0.5,
+    "eval_loss": 1.5774461030960083,
+    "eval_runtime": 11.5703,
+    "eval_samples_per_second": 2592.844,
+    "eval_steps_per_second": 10.199,
+    "eval_subsequence_accuracy": 0.5
+}

pred_mnli-m.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:6c12ce130d9b77b0700f90171f41ff62ac4f3a344f49e645865d1d1309bcfb99
+size 117908

pred_mnli-mm.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:0e078e0eeb8ed9619b7c907b5356d27d7dda1596142d45cc0fc1cd44e2d8a8b8
+size 118112

pred_ood.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:554cc791da745b2bd3ca61805735599a77df9a190f89c861ac2202551eeb04a3
+size 360128

pred_stress-antonym.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:3634e1cd02558ef565b17cec4f06b5666dc85f403303bd44ff230c1a402bd15f
+size 18860

pred_stress-length_mismatch.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:19fec0fb33d8ee1393556c45187d3bd77f416f9bb69204554d9465c31d85ab5d
+size 117908

pred_stress-negation.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:cc09ec8ddbafe19e99224fdb9ddc0187331a9add6ab00aa3bba1201d239c2a69
+size 118112

pred_stress-numerical_reasoning.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:40050cd19e645bffbe87734c1a7654584f984106a49a818df0f3a144f569c52f
+size 91280

pred_stress-spelling_error.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:1a9a7a56b3a3e57039ddb54abc3771f7e6ae4756482fdadb34a9b9c7c37cbe63
+size 109772

pred_stress-word_overlap.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:69389bba28b501f2e98a2fc4d5ccfce3e0deb7a642e0e51f58c3937e019c1cd6
+size 118112

pred_transfer-qnli.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:531819cba4de32a5ba4bf53faa5cf4f0c0a3cf5b8f24db7c57ae7e8163cfb440
+size 65684

pred_transfer-snli.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:3c70523b9de79d5fefb2b093e59a713214bc864cd6df3193e2811544820bef3a
+size 118232

pred_transfer.npz.npy ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a183f763fa480651733ecd9425e39da18e574621c13049029044bc01591252ed
+size 118232

pytorch_model.bin ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a81248d8c31cad16ad17cfbdffb8cfd5266cc4a574752196204c8deb4e182c57
+size 17908563

rep_mnli-m.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:a8a5aa4ac0f4ce65ebc3bf31542d3eb6a623d451586af438f660a8c8220f51e1
+size 10071842

rep_mnli-mm.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:0fe6dfae39207ba7fb6fa599cf117e525c8b60afc2103bc30d058d6239c6df0d
+size 10089332

rep_ood.npz ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:316b5af0250136e96ca504330af8600e580ed11c3496369e1705d368c878dc41
+size 7864386430

rep_ood.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:b40b112b62ab65c501892c811d6044aaa202d8dd2f27a6fd482af460870d3871
+size 30782746

rep_transfer.pt ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:c9a5e59ad64a67a51ecade8e12d351d81559b772ab632f0703aff37bcd82c739
+size 10099654

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": 4.099935938500961,
+    "eval_combined_score": 4.099935938500961,
+    "eval_loss": 2.143801689147949,
+    "eval_runtime": 4.7113,
+    "eval_samples_per_second": 331.328,
+    "eval_steps_per_second": 1.486
+}

stress-length_mismatch_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 60.69281711665817,
+    "eval_combined_score": 60.69281711665817,
+    "eval_loss": 0.8896390795707703,
+    "eval_runtime": 23.8024,
+    "eval_samples_per_second": 412.354,
+    "eval_steps_per_second": 1.638
+}

stress-negation_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 43.480471928397066,
+    "eval_combined_score": 43.480471928397066,
+    "eval_loss": 1.130544662475586,
+    "eval_runtime": 25.2705,
+    "eval_samples_per_second": 389.07,
+    "eval_steps_per_second": 1.543
+}

stress-numerical_reasoning_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 28.001579778830965,
+    "eval_combined_score": 28.001579778830965,
+    "eval_loss": 1.745166301727295,
+    "eval_runtime": 18.6153,
+    "eval_samples_per_second": 408.052,
+    "eval_steps_per_second": 1.612
+}

stress-spelling_error_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 59.363029440735474,
+    "eval_combined_score": 59.363029440735474,
+    "eval_loss": 0.8884789347648621,
+    "eval_runtime": 22.44,
+    "eval_samples_per_second": 407.175,
+    "eval_steps_per_second": 1.604
+}

stress-word_overlap_results.json ADDED Viewed

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

+{
+    "epoch": 3.0,
+    "eval_accuracy": 56.489015459723355,
+    "eval_combined_score": 56.489015459723355,
+    "eval_loss": 0.9394799470901489,
+    "eval_runtime": 23.5578,
+    "eval_samples_per_second": 417.356,
+    "eval_steps_per_second": 1.656
+}

test_results.json ADDED Viewed

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

+{
+    "ave/id": 0.6483877278886412,
+    "ave/ood": 0.5,
+    "ave/transfer": 0.5376955903271693,
+    "gap/constituent": 0.14838772788864119,
+    "gap/lexical_overlap": 0.14838772788864119,
+    "gap/ood": 0.14838772788864119,
+    "gap/subsequence": 0.14838772788864119,
+    "gap/transfer": 0.11069213756147189,
+    "mnli-m": 0.6409577177789099,
+    "mnli-mm": 0.6558177379983726,
+    "ood": 0.5,
+    "ood/constituent": 0.5,
+    "ood/lexical_overlap": 0.5,
+    "ood/subsequence": 0.5,
+    "transfer": 0.5376955903271693
+}

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-2_H-128_A-2",
+  "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": 1.8895709598546233,
+    "train_runtime": 3349.1082,
+    "train_samples_per_second": 351.767,
+    "train_steps_per_second": 10.993
+}

trainer_state.json ADDED Viewed

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

+{
+  "best_metric": 64.09577177789099,
+  "best_model_checkpoint": "outputs/bert-tiny/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": 2.1872,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 2.1413,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 2.1019,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 2.0862,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 2.0576,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 2.0436,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 2.034,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 2.019,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 2.0084,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 2.0018,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 2.0029,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.989,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.9798,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.9742,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.9632,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.9637,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.9591,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.9525,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.9593,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.9482,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.9417,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.9388,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.9328,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.9329,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 62.373917473255226,
+      "eval_combined_score": 62.373917473255226,
+      "eval_loss": 0.8411945104598999,
+      "eval_runtime": 23.4695,
+      "eval_samples_per_second": 418.202,
+      "eval_steps_per_second": 1.662,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 1.9266,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 1.9146,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 1.9165,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 1.9147,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 1.9127,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 1.9101,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 1.8977,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 1.8576,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 1.8791,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 1.8755,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 1.8575,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 1.8536,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 1.8556,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 1.8432,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 1.8495,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 1.845,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 1.8483,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 1.8372,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 1.8484,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 1.8429,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 1.8306,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 1.8354,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 1.8139,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 1.8239,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 1.8289,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 63.65766683647478,
+      "eval_combined_score": 63.65766683647478,
+      "eval_loss": 0.8267493844032288,
+      "eval_runtime": 21.3482,
+      "eval_samples_per_second": 459.757,
+      "eval_steps_per_second": 1.827,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 1.8174,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 1.8226,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 1.7959,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 1.8105,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 1.8114,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 1.8027,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 1.8094,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 1.81,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 1.8028,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.7966,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 1.8043,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 1.8021,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 1.7999,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 1.7899,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 1.8039,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 1.7887,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 1.7902,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 1.8093,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 1.8069,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 1.7928,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 1.7864,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 1.7978,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 1.8001,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 1.8115,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 64.09577177789099,
+      "eval_combined_score": 64.09577177789099,
+      "eval_loss": 0.8214038610458374,
+      "eval_runtime": 20.5305,
+      "eval_samples_per_second": 478.07,
+      "eval_steps_per_second": 1.9,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 1.8895709598546233,
+      "train_runtime": 3349.1082,
+      "train_samples_per_second": 351.767,
+      "train_steps_per_second": 10.993
+    }
+  ],
+  "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:f4629fb2d3e6af86c30211608166274071f8d31b24f1993d0c5587a04db0f229
+size 4091

transfer-qnli_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 51.601684056379284,
+    "eval_combined_score": 51.601684056379284,
+    "eval_loss": 1.589883804321289,
+    "eval_runtime": 9.7098,
+    "eval_samples_per_second": 562.628,
+    "eval_steps_per_second": 2.266
+}

transfer-snli_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 53.73907742328795,
+    "eval_combined_score": 53.73907742328795,
+    "eval_loss": 0.9472026228904724,
+    "eval_runtime": 18.3143,
+    "eval_samples_per_second": 537.393,
+    "eval_steps_per_second": 2.129
+}

transfer_results.json ADDED Viewed

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

+{
+    "eval_accuracy": 0.5376955903271693,
+    "eval_combined_score": 0.5376955903271693,
+    "eval_loss": 0.8987720608711243,
+    "eval_runtime": 3.7795,
+    "eval_samples_per_second": 2604.049,
+    "eval_steps_per_second": 10.319
+}

vocab.txt ADDED Viewed

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