Upload folder using huggingface_hub

all_results.json

@@ -0,0 +1,41 @@
+{
+    "ave/id": 0.41253954359889033,
+    "ave/ood": 0.49596666666666667,
+    "ave/stress": 45.304564189231485,
+    "ave/transfer": 0.37421255842308476,
+    "epoch": 3.0,
+    "eval_accuracy": 0.37421255842308476,
+    "eval_combined_score": 0.37421255842308476,
+    "eval_constituent_accuracy": 0.5014,
+    "eval_lexical_overlap_accuracy": 0.507,
+    "eval_loss": 1.078484058380127,
+    "eval_runtime": 4.4655,
+    "eval_samples_per_second": 2203.995,
+    "eval_steps_per_second": 8.734,
+    "eval_subsequence_accuracy": 0.4795,
+    "gap/constituent": -0.08886045640110962,
+    "gap/lexical_overlap": -0.09446045640110967,
+    "gap/ood": -0.08342712306777633,
+    "gap/stress": -4.050609829342449,
+    "gap/subsequence": -0.06696045640110965,
+    "gap/transfer": 0.03832698517580557,
+    "mnli-m": 0.42221090168110037,
+    "mnli-mm": 0.40286818551668024,
+    "ood": 0.49596666666666667,
+    "ood/constituent": 0.5014,
+    "ood/lexical_overlap": 0.507,
+    "ood/subsequence": 0.4795,
+    "stress-antonym": 72.19730941704036,
+    "stress-length_mismatch": 39.887926642893525,
+    "stress-negation": 40.44955248169243,
+    "stress-numerical_reasoning": 34.55766192733017,
+    "stress-spelling_error": 43.329320345846554,
+    "stress-word_overlap": 41.40561432058584,
+    "train_loss": 1.0340555465413093,
+    "train_runtime": 2034.3039,
+    "train_samples_per_second": 579.12,
+    "train_steps_per_second": 18.098,
+    "transfer": 0.37421255842308476,
+    "transfer-qnli": 51.47354933186894,
+    "transfer-snli": 37.421255842308476
+}

mnli-m_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.42221090168110037,
+    "eval_combined_score": 0.42221090168110037,
+    "eval_loss": 1.0493260622024536,
+    "eval_runtime": 4.0337,
+    "eval_samples_per_second": 2433.231,
+    "eval_steps_per_second": 9.668
+}

mnli-mm_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.40286818551668024,
+    "eval_combined_score": 0.40286818551668024,
+    "eval_loss": 1.0537621974945068,
+    "eval_runtime": 4.5413,
+    "eval_samples_per_second": 2165.028,
+    "eval_steps_per_second": 8.588
+}

ood_results.json

@@ -0,0 +1,11 @@
+{
+    "eval_accuracy": 0.49596666666666667,
+    "eval_combined_score": 0.49596666666666667,
+    "eval_constituent_accuracy": 0.5014,
+    "eval_lexical_overlap_accuracy": 0.507,
+    "eval_loss": 1.3018536567687988,
+    "eval_runtime": 9.5735,
+    "eval_samples_per_second": 3133.636,
+    "eval_steps_per_second": 12.326,
+    "eval_subsequence_accuracy": 0.4795
+}

pred_mnli-m.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73bb93857a18a7c35a0483801b428e4331e4615b8f2aad673b982615769e9814
+size 117908

pred_mnli-mm.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49e63300932076395ba671f736a83c441591bb0f60838faf2e3e61738d433512
+size 118112

pred_ood.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41447f45dca8a84cedb4c82f2f11628859614926aae4949e03c26074413eabe0
+size 360128

pred_stress-antonym.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb2cc4877f8316cdc8e795f538323ddbf1d1a9c0ecb5e7e45f64d0eced5c667d
+size 18860

pred_stress-length_mismatch.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63ba244aa217772551a500a1ebe6291aaa7cec776a16ce55a74d26a6bc50eb00
+size 117908

pred_stress-negation.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:018bb841d366bf55edba12c5e19d42551bf1aa290d4cee6e3852953b96453d69
+size 118112

pred_stress-numerical_reasoning.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db1a035c745952fb0ea0ced57b9bfe69401e9d149d3f2adf2df954c3eaaa8660
+size 91280

pred_stress-spelling_error.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90c2b7ca55c8c7bc3cdc7a7847db1291d2432d5dada35b06fda5192dc83d64d
+size 109772

pred_stress-word_overlap.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1a39e9ffe631009c69a2349d55ccf4bba7ffb03fd1be9652380e08077a36aa
+size 118112

pred_transfer-qnli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71f2aa1d1d37f3bd997a358f73fde5807e1d086c8a26c4a5f336ae60f7d8b6cb
+size 65684

pred_transfer-snli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab1196305a4170c6f19ae241e31e8b574d4290abc406f4fd1e145704417501e
+size 118232

pred_transfer.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab1196305a4170c6f19ae241e31e8b574d4290abc406f4fd1e145704417501e
+size 118232

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:612a47f8f92ccab526f32272c7aa1f7d7251e275414d7ab018d548e819c1a455
+size 35133741

rep_mnli-m.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5f745afa00e8c6ed1340a0a123f21f5bfd850ac6229158902710cbc27418275
+size 10071842

rep_mnli-mm.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c586a36765f4cc3ff6677eef9bcf178f903de05f7c55ae9adcf6308393036c1
+size 10089332

rep_ood.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fc692323dafc7165978177844299178ee88894e603d22c87b08284f9851dbe5
+size 30782746

rep_transfer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f97ee94898d406c33472a1b17a14e38056b0477e916406fa2b2b081ae9832ac5
+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 +1,8 @@
+{
+    "eval_accuracy": 72.19730941704036,
+    "eval_combined_score": 72.19730941704036,
+    "eval_loss": 0.9308596253395081,
+    "eval_runtime": 2.4922,
+    "eval_samples_per_second": 626.362,
+    "eval_steps_per_second": 2.809
+}

stress-length_mismatch_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 39.887926642893525,
+    "eval_combined_score": 39.887926642893525,
+    "eval_loss": 1.0573346614837646,
+    "eval_runtime": 18.1548,
+    "eval_samples_per_second": 540.629,
+    "eval_steps_per_second": 2.148
+}

stress-negation_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 40.44955248169243,
+    "eval_combined_score": 40.44955248169243,
+    "eval_loss": 1.0657862424850464,
+    "eval_runtime": 17.2614,
+    "eval_samples_per_second": 569.595,
+    "eval_steps_per_second": 2.259
+}

stress-numerical_reasoning_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 34.55766192733017,
+    "eval_combined_score": 34.55766192733017,
+    "eval_loss": 1.1387479305267334,
+    "eval_runtime": 13.7937,
+    "eval_samples_per_second": 550.687,
+    "eval_steps_per_second": 2.175
+}

stress-spelling_error_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 43.329320345846554,
+    "eval_combined_score": 43.329320345846554,
+    "eval_loss": 1.049009919166565,
+    "eval_runtime": 18.4532,
+    "eval_samples_per_second": 495.143,
+    "eval_steps_per_second": 1.951
+}

stress-word_overlap_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 41.40561432058584,
+    "eval_combined_score": 41.40561432058584,
+    "eval_loss": 1.0677262544631958,
+    "eval_runtime": 16.2255,
+    "eval_samples_per_second": 605.96,
+    "eval_steps_per_second": 2.404
+}

test_results.json

@@ -0,0 +1,17 @@
+{
+    "ave/id": 0.41253954359889033,
+    "ave/ood": 0.49596666666666667,
+    "ave/transfer": 0.37421255842308476,
+    "gap/constituent": -0.08886045640110962,
+    "gap/lexical_overlap": -0.09446045640110967,
+    "gap/ood": -0.08342712306777633,
+    "gap/subsequence": -0.06696045640110965,
+    "gap/transfer": 0.03832698517580557,
+    "mnli-m": 0.42221090168110037,
+    "mnli-mm": 0.40286818551668024,
+    "ood": 0.49596666666666667,
+    "ood/constituent": 0.5014,
+    "ood/lexical_overlap": 0.507,
+    "ood/subsequence": 0.4795,
+    "transfer": 0.37421255842308476
+}

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": 1.0340555465413093,
+    "train_runtime": 2034.3039,
+    "train_samples_per_second": 579.12,
+    "train_steps_per_second": 18.098
+}

trainer_state.json

@@ -0,0 +1,493 @@
+{
+  "best_metric": 42.22109016811004,
+  "best_model_checkpoint": "outputs/bert-tiny/weak_tiny_poe/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": 1.0659,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 1.0457,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 1.0488,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 1.0466,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 1.0515,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 1.046,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 1.0456,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 1.0397,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 1.0387,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 1.0398,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 1.0423,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.0371,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.0359,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.0356,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.0198,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.0412,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.0364,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.0339,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.0389,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.0347,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.0266,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.0285,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.0283,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.0424,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 41.72185430463576,
+      "eval_combined_score": 41.72185430463576,
+      "eval_loss": 1.0730369091033936,
+      "eval_runtime": 18.0041,
+      "eval_samples_per_second": 545.153,
+      "eval_steps_per_second": 2.166,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 1.0334,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 1.0326,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 1.0336,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 1.0362,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 1.0361,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 1.0387,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 1.0369,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 1.0206,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 1.0431,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 1.0288,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 1.0343,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 1.0234,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 1.037,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 1.0183,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 1.036,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 1.035,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 1.0353,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 1.0221,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 1.0414,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 1.034,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 1.0459,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 1.0369,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 1.0265,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 1.0263,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 1.044,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 39.98981151299032,
+      "eval_combined_score": 39.98981151299032,
+      "eval_loss": 1.0613842010498047,
+      "eval_runtime": 19.4469,
+      "eval_samples_per_second": 504.707,
+      "eval_steps_per_second": 2.005,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 1.0272,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 1.0327,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 1.0186,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 1.024,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 1.0392,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 1.0315,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 1.0301,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 1.0311,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 1.0158,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.0198,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 1.0238,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 1.0321,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 1.0284,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 1.0205,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 1.0382,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 1.0243,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 1.0288,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 1.0421,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 1.0316,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 1.0328,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 1.0289,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 1.024,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 1.032,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 1.0443,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 42.22109016811004,
+      "eval_combined_score": 42.22109016811004,
+      "eval_loss": 1.0493260622024536,
+      "eval_runtime": 17.4192,
+      "eval_samples_per_second": 563.458,
+      "eval_steps_per_second": 2.239,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 1.0340555465413093,
+      "train_runtime": 2034.3039,
+      "train_samples_per_second": 579.12,
+      "train_steps_per_second": 18.098
+    }
+  ],
+  "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:d5425eb84baba82ea359c9ba2e3bb4c66a55d6a6508591d92d74b9ef2d9c4204
+size 4091

transfer-qnli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 51.47354933186894,
+    "eval_combined_score": 51.47354933186894,
+    "eval_loss": 1.063218116760254,
+    "eval_runtime": 9.6691,
+    "eval_samples_per_second": 564.994,
+    "eval_steps_per_second": 2.275
+}

transfer-snli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 37.421255842308476,
+    "eval_combined_score": 37.421255842308476,
+    "eval_loss": 1.078484058380127,
+    "eval_runtime": 18.5894,
+    "eval_samples_per_second": 529.442,
+    "eval_steps_per_second": 2.098
+}

transfer_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.37421255842308476,
+    "eval_combined_score": 0.37421255842308476,
+    "eval_loss": 1.078484058380127,
+    "eval_runtime": 4.4655,
+    "eval_samples_per_second": 2203.995,
+    "eval_steps_per_second": 8.734
+}