Upload folder using huggingface_hub

all_results.json

@@ -0,0 +1,41 @@
+{
+    "ave/id": 0.6455390913759722,
+    "ave/ood": 0.5,
+    "ave/stress": 42.18187896672962,
+    "ave/transfer": 0.535663483031904,
+    "epoch": 3.0,
+    "eval_accuracy": 0.535663483031904,
+    "eval_combined_score": 0.535663483031904,
+    "eval_constituent_accuracy": 0.5,
+    "eval_lexical_overlap_accuracy": 0.5,
+    "eval_loss": 0.9032159447669983,
+    "eval_runtime": 4.4952,
+    "eval_samples_per_second": 2189.427,
+    "eval_steps_per_second": 8.676,
+    "eval_subsequence_accuracy": 0.5,
+    "gap/constituent": 0.1455390913759722,
+    "gap/lexical_overlap": 0.1455390913759722,
+    "gap/ood": 0.1455390913759722,
+    "gap/stress": 22.372030170867596,
+    "gap/subsequence": 0.1455390913759722,
+    "gap/transfer": 0.10987560834406818,
+    "mnli-m": 0.6400407539480387,
+    "mnli-mm": 0.6510374288039056,
+    "ood": 0.5,
+    "ood/constituent": 0.5,
+    "ood/lexical_overlap": 0.5,
+    "ood/subsequence": 0.5,
+    "stress-antonym": 4.932735426008969,
+    "stress-length_mismatch": 60.86602139582272,
+    "stress-negation": 42.07689178193653,
+    "stress-numerical_reasoning": 27.040547656661403,
+    "stress-spelling_error": 60.140089744992885,
+    "stress-word_overlap": 58.03498779495525,
+    "train_loss": 1.8908033308802559,
+    "train_runtime": 3355.0312,
+    "train_samples_per_second": 351.146,
+    "train_steps_per_second": 10.973,
+    "transfer": 0.535663483031904,
+    "transfer-qnli": 51.5101592531576,
+    "transfer-snli": 53.94228815281446
+}

mnli-m_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.6400407539480387,
+    "eval_combined_score": 0.6400407539480387,
+    "eval_loss": 0.8160342574119568,
+    "eval_runtime": 3.8669,
+    "eval_samples_per_second": 2538.189,
+    "eval_steps_per_second": 10.086
+}

mnli-mm_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.6510374288039056,
+    "eval_combined_score": 0.6510374288039056,
+    "eval_loss": 0.7971212267875671,
+    "eval_runtime": 4.5373,
+    "eval_samples_per_second": 2166.935,
+    "eval_steps_per_second": 8.595
+}

ood_results.json

@@ -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": 1.5745669603347778,
+    "eval_runtime": 11.7129,
+    "eval_samples_per_second": 2561.281,
+    "eval_steps_per_second": 10.074,
+    "eval_subsequence_accuracy": 0.5
+}

pred_mnli-m.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:745f205f9f8725f148846c8d2ff0526486dc473b687f23e8f3745570e784b248
+size 117908

pred_mnli-mm.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dd43737915a765120c3bd3af286df8dd16593c2f50118def35ddc16a72bdb4b
+size 118112

pred_ood.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a3f7aec79bea53403e56e312e47ba2723bab134a6e820308efea545b83aa5df
+size 360128

pred_stress-antonym.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89a23353309772d4ab147e5cd64fd5fb693a03d39b8039473a81e72d77d6a43d
+size 18860

pred_stress-length_mismatch.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd605df83be4889cf26e78c0ab4a4737600a1a1f1938b8f983b2e82824aaa446
+size 117908

pred_stress-negation.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ef0ce535289da5e18e906a3f7dcf030aeee8855dab053b344055925453fd61e
+size 118112

pred_stress-numerical_reasoning.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6242ba5d027d14566e2d8fc88fd8964b63d6f068b28e993096d92ebb9c1cb4da
+size 91280

pred_stress-spelling_error.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:196018ba9f383deb52f4c59e59f174e4eddc6219dd741dffc91e505a22e4347e
+size 109772

pred_stress-word_overlap.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9c90833e08673fd12f83f060478273b7509240a3f25afdb39f69b5c01d46ea6
+size 118112

pred_transfer-qnli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8647f98535914ee412fb6413dcb65a13aca95dd7dfd4ee262a5cf6c0f0be7fce
+size 65684

pred_transfer-snli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55ab89b8b1db2b23a604d9eb7832bc834754dcb751423c54c8d3f6d7dfbf3c78
+size 118232

pred_transfer.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:300d890aaf5ab218c38654870a65b96a470712be85aedb81d072a4196527fec3
+size 118232

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab8e1a7383de911f4356d4d22b8eb81d73af637a4c1e41c5ad56af095f5ac4f9
+size 17908563

rep_mnli-m.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5801f41d67010819d4a7395c631254a2f9cacaa6d5de257c7d466adabca26817
+size 10071842

rep_mnli-mm.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d32268b39e80b2200d4a2db44211ee943c1c4738be59db7626f9dd265d3d6f96
+size 10089332

rep_ood.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:316b5af0250136e96ca504330af8600e580ed11c3496369e1705d368c878dc41
+size 7864386430

rep_ood.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ef16b602f3c5614c3cf2c40fd2a736b061a1478e6f247b85447c2bdae8b8660
+size 30782746

rep_transfer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c1a17869d6a67d0beacf12f73b11ca4f6910838c1ab7fb029518d27ed55fb8a
+size 10099654

special_tokens_map.json

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

stress-antonym_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 4.932735426008969,
+    "eval_combined_score": 4.932735426008969,
+    "eval_loss": 2.1483373641967773,
+    "eval_runtime": 4.8314,
+    "eval_samples_per_second": 323.093,
+    "eval_steps_per_second": 1.449
+}

stress-length_mismatch_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 60.86602139582272,
+    "eval_combined_score": 60.86602139582272,
+    "eval_loss": 0.8796690702438354,
+    "eval_runtime": 23.6968,
+    "eval_samples_per_second": 414.191,
+    "eval_steps_per_second": 1.646
+}

stress-negation_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 42.07689178193653,
+    "eval_combined_score": 42.07689178193653,
+    "eval_loss": 1.2378135919570923,
+    "eval_runtime": 25.0333,
+    "eval_samples_per_second": 392.756,
+    "eval_steps_per_second": 1.558
+}

stress-numerical_reasoning_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 27.040547656661403,
+    "eval_combined_score": 27.040547656661403,
+    "eval_loss": 1.7636290788650513,
+    "eval_runtime": 18.59,
+    "eval_samples_per_second": 408.606,
+    "eval_steps_per_second": 1.614
+}

stress-spelling_error_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 60.140089744992885,
+    "eval_combined_score": 60.140089744992885,
+    "eval_loss": 0.8862730860710144,
+    "eval_runtime": 22.9907,
+    "eval_samples_per_second": 397.421,
+    "eval_steps_per_second": 1.566
+}

stress-word_overlap_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 58.03498779495525,
+    "eval_combined_score": 58.03498779495525,
+    "eval_loss": 0.9075659513473511,
+    "eval_runtime": 23.1678,
+    "eval_samples_per_second": 424.381,
+    "eval_steps_per_second": 1.683
+}

test_results.json

@@ -0,0 +1,17 @@
+{
+    "ave/id": 0.6455390913759722,
+    "ave/ood": 0.5,
+    "ave/transfer": 0.535663483031904,
+    "gap/constituent": 0.1455390913759722,
+    "gap/lexical_overlap": 0.1455390913759722,
+    "gap/ood": 0.1455390913759722,
+    "gap/subsequence": 0.1455390913759722,
+    "gap/transfer": 0.10987560834406818,
+    "mnli-m": 0.6400407539480387,
+    "mnli-mm": 0.6510374288039056,
+    "ood": 0.5,
+    "ood/constituent": 0.5,
+    "ood/lexical_overlap": 0.5,
+    "ood/subsequence": 0.5,
+    "transfer": 0.535663483031904
+}

tokenizer_config.json

@@ -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

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.8908033308802559,
+    "train_runtime": 3355.0312,
+    "train_samples_per_second": 351.146,
+    "train_steps_per_second": 10.973
+}

trainer_state.json

@@ -0,0 +1,493 @@
+{
+  "best_metric": 64.00407539480388,
+  "best_model_checkpoint": "outputs/bert-tiny/read/mnli_87/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": 2.1815,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 2.1387,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 2.0997,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 2.0643,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 2.0457,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 2.0252,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 2.0068,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 1.9931,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 1.9752,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 1.9752,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 1.9679,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.9604,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.9559,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.9575,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.9553,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.9591,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.9473,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.9395,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.9363,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.9398,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.9333,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.9329,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.9346,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.9324,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 62.302598064187464,
+      "eval_combined_score": 62.302598064187464,
+      "eval_loss": 0.844679594039917,
+      "eval_runtime": 21.5908,
+      "eval_samples_per_second": 454.591,
+      "eval_steps_per_second": 1.806,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 1.905,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 1.9127,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 1.9104,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 1.9041,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 1.9013,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 1.8899,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 1.9136,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 1.8881,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 1.9068,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 1.8893,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 1.8913,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 1.901,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 1.8843,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 1.887,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 1.883,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 1.8701,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 1.871,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 1.8605,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 1.853,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 1.843,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 1.8527,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 1.8306,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 1.831,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 1.8235,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 1.8396,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 63.71879775853285,
+      "eval_combined_score": 63.71879775853285,
+      "eval_loss": 0.8217228055000305,
+      "eval_runtime": 21.3295,
+      "eval_samples_per_second": 460.16,
+      "eval_steps_per_second": 1.828,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 1.8107,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 1.8124,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 1.8153,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 1.8259,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 1.8101,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 1.8051,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 1.8108,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 1.8217,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 1.8045,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.7932,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 1.8144,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 1.8154,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 1.8175,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 1.8188,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 1.7821,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 1.8137,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 1.8063,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 1.7892,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 1.8049,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 1.8035,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 1.8013,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 1.799,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 1.802,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 1.8052,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 64.00407539480388,
+      "eval_combined_score": 64.00407539480388,
+      "eval_loss": 0.8160342574119568,
+      "eval_runtime": 20.9475,
+      "eval_samples_per_second": 468.552,
+      "eval_steps_per_second": 1.862,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 1.8908033308802559,
+      "train_runtime": 3355.0312,
+      "train_samples_per_second": 351.146,
+      "train_steps_per_second": 10.973
+    }
+  ],
+  "max_steps": 36816,
+  "num_train_epochs": 3,
+  "total_flos": 0.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:336d43cbb7872309d0c020a02d42260d616aa9ee05ebf3114c862c4442c9ddad
+size 4091

transfer-qnli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 51.5101592531576,
+    "eval_combined_score": 51.5101592531576,
+    "eval_loss": 1.5651824474334717,
+    "eval_runtime": 9.8653,
+    "eval_samples_per_second": 553.759,
+    "eval_steps_per_second": 2.23
+}

transfer-snli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 53.94228815281446,
+    "eval_combined_score": 53.94228815281446,
+    "eval_loss": 0.9345145225524902,
+    "eval_runtime": 21.3199,
+    "eval_samples_per_second": 461.635,
+    "eval_steps_per_second": 1.829
+}

transfer_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.535663483031904,
+    "eval_combined_score": 0.535663483031904,
+    "eval_loss": 0.9032159447669983,
+    "eval_runtime": 4.4952,
+    "eval_samples_per_second": 2189.427,
+    "eval_steps_per_second": 8.676
+}