Upload folder using huggingface_hub

all_results.json

@@ -0,0 +1,41 @@
+{
+    "ave/id": 0.4267390064442802,
+    "ave/ood": 0.49366666666666664,
+    "ave/stress": 43.655327788176436,
+    "ave/transfer": 0.40256045519203415,
+    "epoch": 3.0,
+    "eval_accuracy": 0.40256045519203415,
+    "eval_combined_score": 0.40256045519203415,
+    "eval_constituent_accuracy": 0.4954,
+    "eval_lexical_overlap_accuracy": 0.5017,
+    "eval_loss": 1.0609203577041626,
+    "eval_runtime": 3.975,
+    "eval_samples_per_second": 2475.977,
+    "eval_steps_per_second": 9.811,
+    "eval_subsequence_accuracy": 0.4839,
+    "gap/constituent": -0.06866099355571981,
+    "gap/lexical_overlap": -0.07496099355571983,
+    "gap/ood": -0.06692766022238644,
+    "gap/stress": -0.9814271437484194,
+    "gap/subsequence": -0.0571609935557198,
+    "gap/transfer": 0.02417855125224605,
+    "mnli-m": 0.435048395313296,
+    "mnli-mm": 0.4184296175752644,
+    "ood": 0.49366666666666664,
+    "ood/constituent": 0.4954,
+    "ood/lexical_overlap": 0.5017,
+    "ood/subsequence": 0.4839,
+    "stress-antonym": 58.167841127482376,
+    "stress-length_mismatch": 43.15843097300051,
+    "stress-negation": 41.53783563873068,
+    "stress-numerical_reasoning": 33.4913112164297,
+    "stress-spelling_error": 44.58793914851702,
+    "stress-word_overlap": 40.98860862489829,
+    "train_loss": 1.0361233511887027,
+    "train_runtime": 2033.3359,
+    "train_samples_per_second": 579.396,
+    "train_steps_per_second": 18.106,
+    "transfer": 0.40256045519203415,
+    "transfer-qnli": 51.23558484349259,
+    "transfer-snli": 40.25604551920342
+}

mnli-m_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.435048395313296,
+    "eval_combined_score": 0.435048395313296,
+    "eval_loss": 1.0391069650650024,
+    "eval_runtime": 3.9541,
+    "eval_samples_per_second": 2482.218,
+    "eval_steps_per_second": 9.863
+}

mnli-mm_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.4184296175752644,
+    "eval_combined_score": 0.4184296175752644,
+    "eval_loss": 1.043450117111206,
+    "eval_runtime": 3.7692,
+    "eval_samples_per_second": 2608.521,
+    "eval_steps_per_second": 10.347
+}

ood_results.json

@@ -0,0 +1,11 @@
+{
+    "eval_accuracy": 0.49366666666666664,
+    "eval_combined_score": 0.49366666666666664,
+    "eval_constituent_accuracy": 0.4954,
+    "eval_lexical_overlap_accuracy": 0.5017,
+    "eval_loss": 1.336901307106018,
+    "eval_runtime": 10.8469,
+    "eval_samples_per_second": 2765.768,
+    "eval_steps_per_second": 10.879,
+    "eval_subsequence_accuracy": 0.4839
+}

pred_mnli-m.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76add562d24a71484e5f8e0edb7c8b4d18f7f260d8401c0b858532fbac77aa3f
+size 117908

pred_mnli-mm.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b42e2534e0ddce6d73c4e450090f9aef54a7c770a25ff2e9ffaf50715082668
+size 118112

pred_ood.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8937d3f52e51534922dc9a5c9091393e93535204ecd0afd537921f2895900b8e
+size 360128

pred_stress-antonym.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a542a5d689fb6f3baef572e72e5a96aaa5d33c821e49b67a3664b04d64b4d8e
+size 18860

pred_stress-length_mismatch.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23226d7a8d08705491ac149c7bc31566fea1df216080c29c4728043d9c2a22d8
+size 117908

pred_stress-negation.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0450f23a83574f52a2792a7ceaf8e0d858b5b3a7e07f0b25784f40410a72738
+size 118112

pred_stress-numerical_reasoning.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5a826f1bbf64d842732e355caced7c2b7b825d03b3461634b45e4917a624129
+size 91280

pred_stress-spelling_error.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c67f8b7529ec53b721b36f03c99c0693a6cb335c933357b675db69c5f558fb0
+size 109772

pred_stress-word_overlap.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9756307866bc898d30f76d3b895b0e5212b03a5b1fb8a8a48ef201f07d361bc
+size 118112

pred_transfer-qnli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05d55268e698275d19e58089bfb8e50f3be056764985cf7e0cd7e9e02c8cc28f
+size 65684

pred_transfer-snli.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f6ab89a38efdacc0c116a9da027ed941b37f9d905f760bf67151c5e263e5dde
+size 118232

pred_transfer.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f6ab89a38efdacc0c116a9da027ed941b37f9d905f760bf67151c5e263e5dde
+size 118232

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c131ff6d386e4490d84d689829ff531292f454c21e9d6b8631b21fe31c75d272
+size 35133741

rep_mnli-m.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30d682cc68958fa9a59e84c9ce5594a09a740ea5076fb2f1585d7ac6a5c98a02
+size 10071842

rep_mnli-mm.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7aeec1b1caef4a84a197fbd9645cc0eb68b3f8431ed5ed61af020d2f133e90ac
+size 10089332

rep_ood.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a06d3a6ffe2ed2a496b025b9f181277618b06a15ae0dae0ea448c02a25a4f81
+size 30782746

rep_transfer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fca3d429e64f61803abf279e486107e92f5804f6e4c014e6a7eba4aa285a9b4
+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": 58.167841127482376,
+    "eval_combined_score": 58.167841127482376,
+    "eval_loss": 0.9379172921180725,
+    "eval_runtime": 2.1552,
+    "eval_samples_per_second": 724.309,
+    "eval_steps_per_second": 3.248
+}

stress-length_mismatch_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 43.15843097300051,
+    "eval_combined_score": 43.15843097300051,
+    "eval_loss": 1.051645040512085,
+    "eval_runtime": 13.5175,
+    "eval_samples_per_second": 726.094,
+    "eval_steps_per_second": 2.885
+}

stress-negation_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 41.53783563873068,
+    "eval_combined_score": 41.53783563873068,
+    "eval_loss": 1.060198187828064,
+    "eval_runtime": 13.6529,
+    "eval_samples_per_second": 720.138,
+    "eval_steps_per_second": 2.857
+}

stress-numerical_reasoning_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 33.4913112164297,
+    "eval_combined_score": 33.4913112164297,
+    "eval_loss": 1.1366249322891235,
+    "eval_runtime": 10.4993,
+    "eval_samples_per_second": 723.48,
+    "eval_steps_per_second": 2.857
+}

stress-spelling_error_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 44.58793914851702,
+    "eval_combined_score": 44.58793914851702,
+    "eval_loss": 1.0406650304794312,
+    "eval_runtime": 12.7253,
+    "eval_samples_per_second": 718.019,
+    "eval_steps_per_second": 2.829
+}

stress-word_overlap_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 40.98860862489829,
+    "eval_combined_score": 40.98860862489829,
+    "eval_loss": 1.0662919282913208,
+    "eval_runtime": 13.6334,
+    "eval_samples_per_second": 721.168,
+    "eval_steps_per_second": 2.861
+}

test_results.json

@@ -0,0 +1,17 @@
+{
+    "ave/id": 0.4267390064442802,
+    "ave/ood": 0.49366666666666664,
+    "ave/transfer": 0.40256045519203415,
+    "gap/constituent": -0.06866099355571981,
+    "gap/lexical_overlap": -0.07496099355571983,
+    "gap/ood": -0.06692766022238644,
+    "gap/subsequence": -0.0571609935557198,
+    "gap/transfer": 0.02417855125224605,
+    "mnli-m": 0.435048395313296,
+    "mnli-mm": 0.4184296175752644,
+    "ood": 0.49366666666666664,
+    "ood/constituent": 0.4954,
+    "ood/lexical_overlap": 0.5017,
+    "ood/subsequence": 0.4839,
+    "transfer": 0.40256045519203415
+}

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": 1.0361233511887027,
+    "train_runtime": 2033.3359,
+    "train_samples_per_second": 579.396,
+    "train_steps_per_second": 18.106
+}

trainer_state.json

@@ -0,0 +1,493 @@
+{
+  "best_metric": 43.504839531329594,
+  "best_model_checkpoint": "outputs/bert-tiny/weak_tiny_poe/mnli_42/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": 1.057,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9456757931334205e-05,
+      "loss": 1.0433,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9185136897001307e-05,
+      "loss": 1.0288,
+      "step": 1500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8913515862668405e-05,
+      "loss": 1.0371,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.8641894828335507e-05,
+      "loss": 1.0506,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.837027379400261e-05,
+      "loss": 1.053,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.809865275966971e-05,
+      "loss": 1.0511,
+      "step": 3500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.7827031725336812e-05,
+      "loss": 1.0417,
+      "step": 4000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.7555410691003914e-05,
+      "loss": 1.0188,
+      "step": 4500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7283789656671015e-05,
+      "loss": 1.0416,
+      "step": 5000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7012168622338114e-05,
+      "loss": 1.0325,
+      "step": 5500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.6740547588005215e-05,
+      "loss": 1.0418,
+      "step": 6000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6468926553672317e-05,
+      "loss": 1.0451,
+      "step": 6500
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.619730551933942e-05,
+      "loss": 1.0524,
+      "step": 7000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.592568448500652e-05,
+      "loss": 1.038,
+      "step": 7500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5654063450673622e-05,
+      "loss": 1.0383,
+      "step": 8000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.5382442416340724e-05,
+      "loss": 1.0431,
+      "step": 8500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.5110821382007822e-05,
+      "loss": 1.0342,
+      "step": 9000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4839200347674924e-05,
+      "loss": 1.0363,
+      "step": 9500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4567579313342026e-05,
+      "loss": 1.0312,
+      "step": 10000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.4295958279009128e-05,
+      "loss": 1.0406,
+      "step": 10500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.402433724467623e-05,
+      "loss": 1.0361,
+      "step": 11000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.3752716210343331e-05,
+      "loss": 1.0359,
+      "step": 11500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.3481095176010431e-05,
+      "loss": 1.0391,
+      "step": 12000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 41.01884870096791,
+      "eval_combined_score": 41.01884870096791,
+      "eval_loss": 1.0738756656646729,
+      "eval_runtime": 18.0382,
+      "eval_samples_per_second": 544.124,
+      "eval_steps_per_second": 2.162,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3209474141677531e-05,
+      "loss": 1.0379,
+      "step": 12500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2937853107344633e-05,
+      "loss": 1.0491,
+      "step": 13000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2666232073011735e-05,
+      "loss": 1.0353,
+      "step": 13500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2394611038678836e-05,
+      "loss": 1.0331,
+      "step": 14000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2122990004345938e-05,
+      "loss": 1.0391,
+      "step": 14500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.185136897001304e-05,
+      "loss": 1.0362,
+      "step": 15000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.157974793568014e-05,
+      "loss": 1.0205,
+      "step": 15500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.130812690134724e-05,
+      "loss": 1.0269,
+      "step": 16000
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.1036505867014341e-05,
+      "loss": 1.0285,
+      "step": 16500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0764884832681443e-05,
+      "loss": 1.0456,
+      "step": 17000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0493263798348545e-05,
+      "loss": 1.0226,
+      "step": 17500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0221642764015647e-05,
+      "loss": 1.0418,
+      "step": 18000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.950021729682747e-06,
+      "loss": 1.0332,
+      "step": 18500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.678400695349848e-06,
+      "loss": 1.0427,
+      "step": 19000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.40677966101695e-06,
+      "loss": 1.0379,
+      "step": 19500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.13515862668405e-06,
+      "loss": 1.0382,
+      "step": 20000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.863537592351152e-06,
+      "loss": 1.0333,
+      "step": 20500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.591916558018254e-06,
+      "loss": 1.0387,
+      "step": 21000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 8.320295523685355e-06,
+      "loss": 1.0436,
+      "step": 21500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 8.048674489352455e-06,
+      "loss": 1.0339,
+      "step": 22000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.777053455019557e-06,
+      "loss": 1.032,
+      "step": 22500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.505432420686659e-06,
+      "loss": 1.0411,
+      "step": 23000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 7.23381138635376e-06,
+      "loss": 1.0244,
+      "step": 23500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.962190352020861e-06,
+      "loss": 1.0385,
+      "step": 24000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.690569317687963e-06,
+      "loss": 1.042,
+      "step": 24500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 41.84411614875191,
+      "eval_combined_score": 41.84411614875191,
+      "eval_loss": 1.051668405532837,
+      "eval_runtime": 19.2677,
+      "eval_samples_per_second": 509.403,
+      "eval_steps_per_second": 2.024,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 6.418948283355063e-06,
+      "loss": 1.0436,
+      "step": 25000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.147327249022165e-06,
+      "loss": 1.0396,
+      "step": 25500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 5.8757062146892665e-06,
+      "loss": 1.0155,
+      "step": 26000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.6040851803563665e-06,
+      "loss": 1.0258,
+      "step": 26500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.332464146023468e-06,
+      "loss": 1.028,
+      "step": 27000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.06084311169057e-06,
+      "loss": 1.0289,
+      "step": 27500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.789222077357671e-06,
+      "loss": 1.036,
+      "step": 28000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.5176010430247726e-06,
+      "loss": 1.0242,
+      "step": 28500
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.2459800086918734e-06,
+      "loss": 1.0335,
+      "step": 29000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.0435,
+      "step": 29500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.702737940026076e-06,
+      "loss": 1.0365,
+      "step": 30000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.431116905693177e-06,
+      "loss": 1.0346,
+      "step": 30500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 3.1594958713602786e-06,
+      "loss": 1.029,
+      "step": 31000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.8878748370273795e-06,
+      "loss": 1.0362,
+      "step": 31500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.6162538026944812e-06,
+      "loss": 1.0327,
+      "step": 32000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.344632768361582e-06,
+      "loss": 1.0256,
+      "step": 32500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.0730117340286834e-06,
+      "loss": 1.0257,
+      "step": 33000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.8013906996957845e-06,
+      "loss": 1.0116,
+      "step": 33500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5297696653628858e-06,
+      "loss": 1.0318,
+      "step": 34000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.258148631029987e-06,
+      "loss": 1.0411,
+      "step": 34500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.865275966970884e-07,
+      "loss": 1.036,
+      "step": 35000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.149065623641896e-07,
+      "loss": 1.0348,
+      "step": 35500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.432855280312908e-07,
+      "loss": 1.0467,
+      "step": 36000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 1.7166449369839202e-07,
+      "loss": 1.0305,
+      "step": 36500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 43.504839531329594,
+      "eval_combined_score": 43.504839531329594,
+      "eval_loss": 1.0391069650650024,
+      "eval_runtime": 17.8035,
+      "eval_samples_per_second": 551.296,
+      "eval_steps_per_second": 2.191,
+      "step": 36816
+    },
+    {
+      "epoch": 3.0,
+      "step": 36816,
+      "total_flos": 0.0,
+      "train_loss": 1.0361233511887027,
+      "train_runtime": 2033.3359,
+      "train_samples_per_second": 579.396,
+      "train_steps_per_second": 18.106
+    }
+  ],
+  "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:cb513dc400f90a54266aab86b7863ee699c3fb4e5f7f48a3efa4fd2bb3a599d5
+size 4091

transfer-qnli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 51.23558484349259,
+    "eval_combined_score": 51.23558484349259,
+    "eval_loss": 1.1406018733978271,
+    "eval_runtime": 9.6672,
+    "eval_samples_per_second": 565.109,
+    "eval_steps_per_second": 2.276
+}

transfer-snli_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 40.25604551920342,
+    "eval_combined_score": 40.25604551920342,
+    "eval_loss": 1.0609203577041626,
+    "eval_runtime": 18.9534,
+    "eval_samples_per_second": 519.273,
+    "eval_steps_per_second": 2.058
+}

transfer_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_accuracy": 0.40256045519203415,
+    "eval_combined_score": 0.40256045519203415,
+    "eval_loss": 1.0609203577041626,
+    "eval_runtime": 3.975,
+    "eval_samples_per_second": 2475.977,
+    "eval_steps_per_second": 9.811
+}