Upload folder using huggingface_hub

all_results.json

@@ -0,0 +1,29 @@
+{
+    "ave/id": 0.7202648869574215,
+    "ave/ood": 0.6133564614050304,
+    "ave/stress": 76.8397970175716,
+    "ave/transfer": 0.8147013782542112,
+    "epoch": 3.0,
+    "eval_accuracy": 70.34313725490196,
+    "eval_combined_score": 0.8147013782542112,
+    "eval_f1": 0.8147013782542112,
+    "eval_loss": 0.7417141795158386,
+    "eval_runtime": 1.0302,
+    "eval_samples_per_second": 396.047,
+    "eval_steps_per_second": 1.941,
+    "gap/ood": 0.1069084255523911,
+    "gap/stress": -4.813308321829439,
+    "gap/transfer": -0.09443649129678966,
+    "ood": 0.6133564614050304,
+    "qqp": 0.7202648869574215,
+    "stress-antonym": 98.87050936384173,
+    "stress-length_mismatch": 70.7109634551495,
+    "stress-negation": 72.69133164151195,
+    "stress-spelling_error": 69.53682797339354,
+    "stress-word_overlap": 72.38935265396125,
+    "train_loss": 0.950482799304254,
+    "train_runtime": 3151.1724,
+    "train_samples_per_second": 346.391,
+    "train_steps_per_second": 10.825,
+    "transfer": 0.8147013782542112
+}

ood_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_combined_score": 0.6133564614050304,
+    "eval_f1": 0.6133564614050304,
+    "eval_loss": 1.627290964126587,
+    "eval_runtime": 3.5206,
+    "eval_samples_per_second": 2272.336,
+    "eval_steps_per_second": 9.089
+}

pred_ood.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1260f432e3e474085be4e6840c043d3091554ab9bf6b0c67fd64bd88487c3888
+size 64128

pred_qqp.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f68d11e49fa2e6a6e096f4d96c5acaa5365a8e5186c0d6feb3c29e406e93ebc
+size 323568

pred_stress-antonym.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55809ffe9e5c570100bed7eb2273476c085a73baefec21de92b52d79d51714bc
+size 55640

pred_stress-length_mismatch.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9baa8046e626df3ae118b6e5a071b1cdacdd38ff7ad385f2f9fadaa698f992f0
+size 323568

pred_stress-negation.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f458c476865214b1ca8de0cbd8f500a327920c2eb013810d770eee87f7bd989e
+size 323568

pred_stress-spelling_error.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb344c68ea31d81dc8b67d49bf7f5921691b9bfc0c28d4fb99d2e425df16d98d
+size 323560

pred_stress-word_overlap.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f5a44194663d846c26900c6bb4ee1a59329e9dd4ffeb717d303a76c25130937
+size 323568

pred_transfer.npz.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d88ad499efb00addb7a2ebc39c9d353f08af7148e0fb3583c9b46e363f43f69
+size 3392

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f52ae1fd202f0ab7d22da42963d40a4316ad4cd21de7410d3445cfa65555fd7a
+size 17907539

qqp_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_combined_score": 0.7202648869574215,
+    "eval_f1": 0.7202648869574215,
+    "eval_loss": 0.5221115350723267,
+    "eval_runtime": 13.1461,
+    "eval_samples_per_second": 3075.444,
+    "eval_steps_per_second": 12.019
+}

rep_ood.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:877862d151cd253d979b4d74cc9db19182f0e4ce309fc07fbaadfd4510b1837e
+size 2097169238

rep_ood.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb07f1cd33343ad1dc065284bc46630d2e4468225872140cf80324d0c4e380d0
+size 8208726

rep_qqp.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25bbb44f6ab1c9a6bb4d0b20282c7cc4e289bba7f3371b10371120428770865a
+size 41484538

rep_transfer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28955c7836a2a7b4867aad73d88ee1eee5414a6432f575401ee7a44a903c41a6
+size 419832

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 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 97.76624873901139,
+    "eval_combined_score": 98.31837905142656,
+    "eval_f1": 98.87050936384173,
+    "eval_loss": 0.15600410103797913,
+    "eval_runtime": 16.709,
+    "eval_samples_per_second": 415.286,
+    "eval_steps_per_second": 1.676
+}

stress-length_mismatch_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 78.1944100915162,
+    "eval_combined_score": 74.45268677333286,
+    "eval_f1": 70.7109634551495,
+    "eval_loss": 0.44404110312461853,
+    "eval_runtime": 95.6539,
+    "eval_samples_per_second": 422.67,
+    "eval_steps_per_second": 1.652
+}

stress-negation_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 74.71432104872619,
+    "eval_combined_score": 73.70282634511906,
+    "eval_f1": 72.69133164151195,
+    "eval_loss": 0.5010897517204285,
+    "eval_runtime": 96.6967,
+    "eval_samples_per_second": 418.111,
+    "eval_steps_per_second": 1.634
+}

stress-spelling_error_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 75.30485542556086,
+    "eval_combined_score": 72.4208416994772,
+    "eval_f1": 69.53682797339354,
+    "eval_loss": 0.4926765561103821,
+    "eval_runtime": 95.7074,
+    "eval_samples_per_second": 422.423,
+    "eval_steps_per_second": 1.651
+}

stress-word_overlap_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 73.98466485283205,
+    "eval_combined_score": 73.18700875339665,
+    "eval_f1": 72.38935265396125,
+    "eval_loss": 0.5098819732666016,
+    "eval_runtime": 96.8935,
+    "eval_samples_per_second": 417.262,
+    "eval_steps_per_second": 1.631
+}

test_results.json

@@ -0,0 +1,10 @@
+{
+    "ave/id": 0.7202648869574215,
+    "ave/ood": 0.6133564614050304,
+    "ave/transfer": 0.8147013782542112,
+    "gap/ood": 0.1069084255523911,
+    "gap/transfer": -0.09443649129678966,
+    "ood": 0.6133564614050304,
+    "qqp": 0.7202648869574215,
+    "transfer": 0.8147013782542112
+}

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": 0.950482799304254,
+    "train_runtime": 3151.1724,
+    "train_samples_per_second": 346.391,
+    "train_steps_per_second": 10.825
+}

trainer_state.json

@@ -0,0 +1,466 @@
+{
+  "best_metric": 72.02648869574216,
+  "best_model_checkpoint": "outputs/bert-tiny/read/qqp_87/checkpoint-34113",
+  "epoch": 3.0,
+  "global_step": 34113,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9706856623574592e-05,
+      "loss": 1.2441,
+      "step": 500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9413713247149183e-05,
+      "loss": 1.1557,
+      "step": 1000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9120569870723774e-05,
+      "loss": 1.1329,
+      "step": 1500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.882742649429836e-05,
+      "loss": 1.1005,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.8534283117872952e-05,
+      "loss": 1.0837,
+      "step": 2500
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8241139741447543e-05,
+      "loss": 1.0757,
+      "step": 3000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.7947996365022134e-05,
+      "loss": 1.0474,
+      "step": 3500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7654852988596725e-05,
+      "loss": 1.0291,
+      "step": 4000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.7361709612171312e-05,
+      "loss": 1.034,
+      "step": 4500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7068566235745903e-05,
+      "loss": 1.03,
+      "step": 5000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.6775422859320497e-05,
+      "loss": 1.0145,
+      "step": 5500
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6482279482895085e-05,
+      "loss": 1.01,
+      "step": 6000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6189136106469676e-05,
+      "loss": 1.0004,
+      "step": 6500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5895992730044266e-05,
+      "loss": 1.0056,
+      "step": 7000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5602849353618857e-05,
+      "loss": 0.9981,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.5309705977193448e-05,
+      "loss": 0.9831,
+      "step": 8000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.5016562600768037e-05,
+      "loss": 0.9832,
+      "step": 8500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4723419224342628e-05,
+      "loss": 0.9904,
+      "step": 9000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.4430275847917217e-05,
+      "loss": 0.9848,
+      "step": 9500
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.4137132471491808e-05,
+      "loss": 0.9755,
+      "step": 10000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.3843989095066399e-05,
+      "loss": 0.9573,
+      "step": 10500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.3550845718640988e-05,
+      "loss": 0.9667,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 67.91244125649271,
+      "eval_combined_score": 68.34303011193873,
+      "eval_f1": 68.77361896738476,
+      "eval_loss": 0.5894622802734375,
+      "eval_runtime": 90.4161,
+      "eval_samples_per_second": 447.155,
+      "eval_steps_per_second": 1.747,
+      "step": 11371
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3257702342215579e-05,
+      "loss": 0.9553,
+      "step": 11500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.2964558965790168e-05,
+      "loss": 0.9386,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.2671415589364759e-05,
+      "loss": 0.9366,
+      "step": 12500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.2378272212939351e-05,
+      "loss": 0.94,
+      "step": 13000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.208512883651394e-05,
+      "loss": 0.9323,
+      "step": 13500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.1791985460088531e-05,
+      "loss": 0.9444,
+      "step": 14000
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.149884208366312e-05,
+      "loss": 0.9268,
+      "step": 14500
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.1205698707237711e-05,
+      "loss": 0.9294,
+      "step": 15000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.0912555330812302e-05,
+      "loss": 0.9281,
+      "step": 15500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.0619411954386891e-05,
+      "loss": 0.9398,
+      "step": 16000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.0326268577961482e-05,
+      "loss": 0.9178,
+      "step": 16500
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0033125201536071e-05,
+      "loss": 0.9188,
+      "step": 17000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.739981825110662e-06,
+      "loss": 0.9254,
+      "step": 17500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.446838448685253e-06,
+      "loss": 0.9184,
+      "step": 18000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.153695072259843e-06,
+      "loss": 0.9133,
+      "step": 18500
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.860551695834433e-06,
+      "loss": 0.928,
+      "step": 19000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.567408319409023e-06,
+      "loss": 0.9055,
+      "step": 19500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 8.274264942983614e-06,
+      "loss": 0.9082,
+      "step": 20000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.981121566558205e-06,
+      "loss": 0.9268,
+      "step": 20500
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.687978190132794e-06,
+      "loss": 0.9281,
+      "step": 21000
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 7.394834813707384e-06,
+      "loss": 0.9294,
+      "step": 21500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.101691437281976e-06,
+      "loss": 0.9226,
+      "step": 22000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.808548060856566e-06,
+      "loss": 0.9154,
+      "step": 22500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 71.36532278011379,
+      "eval_combined_score": 71.09875067982031,
+      "eval_f1": 70.83217857952684,
+      "eval_loss": 0.5470175743103027,
+      "eval_runtime": 89.5769,
+      "eval_samples_per_second": 451.344,
+      "eval_steps_per_second": 1.764,
+      "step": 22742
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.515404684431156e-06,
+      "loss": 0.9096,
+      "step": 23000
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 6.222261308005746e-06,
+      "loss": 0.899,
+      "step": 23500
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 5.929117931580336e-06,
+      "loss": 0.8881,
+      "step": 24000
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 5.6359745551549274e-06,
+      "loss": 0.8952,
+      "step": 24500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 5.3428311787295174e-06,
+      "loss": 0.8906,
+      "step": 25000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 5.049687802304107e-06,
+      "loss": 0.9021,
+      "step": 25500
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 4.756544425878697e-06,
+      "loss": 0.8924,
+      "step": 26000
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.463401049453288e-06,
+      "loss": 0.8812,
+      "step": 26500
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 4.170257673027878e-06,
+      "loss": 0.892,
+      "step": 27000
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 3.877114296602469e-06,
+      "loss": 0.8974,
+      "step": 27500
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 3.583970920177059e-06,
+      "loss": 0.893,
+      "step": 28000
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 3.290827543751649e-06,
+      "loss": 0.8885,
+      "step": 28500
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 2.9976841673262398e-06,
+      "loss": 0.8936,
+      "step": 29000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 2.7045407909008298e-06,
+      "loss": 0.8972,
+      "step": 29500
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 2.41139741447542e-06,
+      "loss": 0.8764,
+      "step": 30000
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 2.1182540380500106e-06,
+      "loss": 0.8957,
+      "step": 30500
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 1.825110661624601e-06,
+      "loss": 0.8902,
+      "step": 31000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.531967285199191e-06,
+      "loss": 0.8869,
+      "step": 31500
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.2388239087737813e-06,
+      "loss": 0.8809,
+      "step": 32000
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 9.456805323483717e-07,
+      "loss": 0.8876,
+      "step": 32500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 6.525371559229619e-07,
+      "loss": 0.9025,
+      "step": 33000
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 3.5939377949755233e-07,
+      "loss": 0.872,
+      "step": 33500
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 6.62504030721426e-08,
+      "loss": 0.8964,
+      "step": 34000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 73.25253524610437,
+      "eval_combined_score": 72.63951197092327,
+      "eval_f1": 72.02648869574216,
+      "eval_loss": 0.5221115350723267,
+      "eval_runtime": 89.4824,
+      "eval_samples_per_second": 451.821,
+      "eval_steps_per_second": 1.766,
+      "step": 34113
+    },
+    {
+      "epoch": 3.0,
+      "step": 34113,
+      "total_flos": 0.0,
+      "train_loss": 0.950482799304254,
+      "train_runtime": 3151.1724,
+      "train_samples_per_second": 346.391,
+      "train_steps_per_second": 10.825
+    }
+  ],
+  "max_steps": 34113,
+  "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 sha256:19b26c5be5cf5528350e46bfe6be2c85bd79b47a1f52c8e8b8b01cd8bbd3c8c8
+size 4091

transfer_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_combined_score": 0.8147013782542112,
+    "eval_f1": 0.8147013782542112,
+    "eval_loss": 0.7417141795158386,
+    "eval_runtime": 1.0302,
+    "eval_samples_per_second": 396.047,
+    "eval_steps_per_second": 1.941
+}