Upload folder using huggingface_hub
Browse files- 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
- special_tokens_map.json +7 -0
- stress-antonym_results.json +8 -0
- stress-length_mismatch_results.json +8 -0
- stress-negation_results.json +8 -0
- stress-numerical_reasoning_results.json +8 -0
- stress-spelling_error_results.json +8 -0
- stress-word_overlap_results.json +8 -0
- test_results.json +17 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- train_results.json +7 -0
- trainer_state.json +967 -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
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"ave/id": 0.6846805289640281,
|
| 3 |
+
"ave/ood": 0.5003333333333333,
|
| 4 |
+
"ave/stress": 44.38108067813665,
|
| 5 |
+
"ave/transfer": 0.5795570006096322,
|
| 6 |
+
"epoch": 6.0,
|
| 7 |
+
"eval_accuracy": 0.5795570006096322,
|
| 8 |
+
"eval_combined_score": 0.5795570006096322,
|
| 9 |
+
"eval_constituent_accuracy": 0.501,
|
| 10 |
+
"eval_lexical_overlap_accuracy": 0.5,
|
| 11 |
+
"eval_loss": 0.8794325590133667,
|
| 12 |
+
"eval_runtime": 3.9813,
|
| 13 |
+
"eval_samples_per_second": 2472.044,
|
| 14 |
+
"eval_steps_per_second": 9.796,
|
| 15 |
+
"eval_subsequence_accuracy": 0.5,
|
| 16 |
+
"gap/constituent": 0.1836805289640281,
|
| 17 |
+
"gap/lexical_overlap": 0.1846805289640281,
|
| 18 |
+
"gap/ood": 0.1843471956306948,
|
| 19 |
+
"gap/stress": 24.086972218266162,
|
| 20 |
+
"gap/subsequence": 0.1846805289640281,
|
| 21 |
+
"gap/transfer": 0.10512352835439587,
|
| 22 |
+
"mnli-m": 0.6798777381558838,
|
| 23 |
+
"mnli-mm": 0.6894833197721725,
|
| 24 |
+
"ood": 0.5003333333333333,
|
| 25 |
+
"ood/constituent": 0.501,
|
| 26 |
+
"ood/lexical_overlap": 0.5,
|
| 27 |
+
"ood/subsequence": 0.5,
|
| 28 |
+
"stress-antonym": 3.9077514413837284,
|
| 29 |
+
"stress-length_mismatch": 64.97198166072339,
|
| 30 |
+
"stress-negation": 44.11106590724166,
|
| 31 |
+
"stress-numerical_reasoning": 26.645602948920484,
|
| 32 |
+
"stress-spelling_error": 64.63828390062383,
|
| 33 |
+
"stress-word_overlap": 62.01179820992677,
|
| 34 |
+
"train_loss": 25.68384859034519,
|
| 35 |
+
"train_runtime": 8782.0618,
|
| 36 |
+
"train_samples_per_second": 268.298,
|
| 37 |
+
"train_steps_per_second": 8.384,
|
| 38 |
+
"transfer": 0.5795570006096322,
|
| 39 |
+
"transfer-qnli": 51.41863444993593,
|
| 40 |
+
"transfer-snli": 57.955700060963224
|
| 41 |
+
}
|
mnli-m_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.6798777381558838,
|
| 3 |
+
"eval_combined_score": 0.6798777381558838,
|
| 4 |
+
"eval_loss": 0.7376300692558289,
|
| 5 |
+
"eval_runtime": 5.5973,
|
| 6 |
+
"eval_samples_per_second": 1753.515,
|
| 7 |
+
"eval_steps_per_second": 6.968
|
| 8 |
+
}
|
mnli-mm_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.6894833197721725,
|
| 3 |
+
"eval_combined_score": 0.6894833197721725,
|
| 4 |
+
"eval_loss": 0.7218872308731079,
|
| 5 |
+
"eval_runtime": 4.8394,
|
| 6 |
+
"eval_samples_per_second": 2031.654,
|
| 7 |
+
"eval_steps_per_second": 8.059
|
| 8 |
+
}
|
ood_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.5003333333333333,
|
| 3 |
+
"eval_combined_score": 0.5003333333333333,
|
| 4 |
+
"eval_constituent_accuracy": 0.501,
|
| 5 |
+
"eval_lexical_overlap_accuracy": 0.5,
|
| 6 |
+
"eval_loss": 1.8080086708068848,
|
| 7 |
+
"eval_runtime": 12.6642,
|
| 8 |
+
"eval_samples_per_second": 2368.885,
|
| 9 |
+
"eval_steps_per_second": 9.318,
|
| 10 |
+
"eval_subsequence_accuracy": 0.5
|
| 11 |
+
}
|
pred_mnli-m.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c5ddc6af3689c94a05c13f3bf59d8d4ccdf69218ea745778d47f428716f68c7
|
| 3 |
+
size 117908
|
pred_mnli-mm.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c00f058b0804bab6bb84d5fcaf308d1c5053da183ad822bf39c978846e26c92
|
| 3 |
+
size 118112
|
pred_ood.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:451a9f940b8f4c1423fde55a56e217eae50b8b63d7412c2c79ef1d336aabecad
|
| 3 |
+
size 360128
|
pred_stress-antonym.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c9701de6c0a54b2c028c04f700f1b773cae5139362aebc7728cec45234e6ff3
|
| 3 |
+
size 18860
|
pred_stress-length_mismatch.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:131b0920ade0a227f6ac2b2a1119235dce0ca126f0f99646f6af733939773796
|
| 3 |
+
size 117908
|
pred_stress-negation.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2cbb8f501f413548e05b413f94e6c4418c4a5a214d7ab84ccf2cb9ab31412ccb
|
| 3 |
+
size 118112
|
pred_stress-numerical_reasoning.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09146ce32fd61db3189cbbde049704e44667aea372c70dc4de2348da253a3059
|
| 3 |
+
size 91280
|
pred_stress-spelling_error.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fb3da90e4533818d5cda6ea503fa9396aaecbc8270f540b8deae30001c5a2e8
|
| 3 |
+
size 109772
|
pred_stress-word_overlap.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2cad405594e1605e1adb5fbbe2308c0e876d53445f9534816adb3ed074cd3e69
|
| 3 |
+
size 118112
|
pred_transfer-qnli.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b352e0691b39e10d9bb4c8ce70c5d7fcdb93e9f480e81088a2acb7ae7c46a420
|
| 3 |
+
size 65684
|
pred_transfer-snli.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c04f025341f5dc47b5d2dec9aacf4629936cf34c4c3551210f3a85fe8877b5b
|
| 3 |
+
size 118232
|
pred_transfer.npz.npy
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c04f025341f5dc47b5d2dec9aacf4629936cf34c4c3551210f3a85fe8877b5b
|
| 3 |
+
size 118232
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:810503c3572155335dd433785a8df6807142728e6d4eba9e01b7c03c0da5bb7b
|
| 3 |
+
size 17586951
|
rep_mnli-m.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bebdd4f164a9c90921a5979d2c476211885104831ecfe071288b2674217269be
|
| 3 |
+
size 10071842
|
rep_mnli-mm.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42ba38c5df4a548665f885c11cf3adb3186f8738c9c688212d75a5b2b7665dc7
|
| 3 |
+
size 10089332
|
rep_ood.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14df12bc490bfaa264f2d40422f025c53cbe29c09a2770e78daed6fd415b7af2
|
| 3 |
+
size 7864386430
|
rep_ood.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbaee8d83de4329ac58a759924c3ddd9ecf90bf8fa0ab8c5d3680c967068499c
|
| 3 |
+
size 30782746
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
stress-antonym_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 3.9077514413837284,
|
| 3 |
+
"eval_combined_score": 3.9077514413837284,
|
| 4 |
+
"eval_loss": 2.7325708866119385,
|
| 5 |
+
"eval_runtime": 2.4997,
|
| 6 |
+
"eval_samples_per_second": 624.476,
|
| 7 |
+
"eval_steps_per_second": 2.8
|
| 8 |
+
}
|
stress-length_mismatch_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 64.97198166072339,
|
| 3 |
+
"eval_combined_score": 64.97198166072339,
|
| 4 |
+
"eval_loss": 0.8164553046226501,
|
| 5 |
+
"eval_runtime": 16.0227,
|
| 6 |
+
"eval_samples_per_second": 612.569,
|
| 7 |
+
"eval_steps_per_second": 2.434
|
| 8 |
+
}
|
stress-negation_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 44.11106590724166,
|
| 3 |
+
"eval_combined_score": 44.11106590724166,
|
| 4 |
+
"eval_loss": 1.2928410768508911,
|
| 5 |
+
"eval_runtime": 15.7558,
|
| 6 |
+
"eval_samples_per_second": 624.024,
|
| 7 |
+
"eval_steps_per_second": 2.475
|
| 8 |
+
}
|
stress-numerical_reasoning_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 26.645602948920484,
|
| 3 |
+
"eval_combined_score": 26.645602948920484,
|
| 4 |
+
"eval_loss": 1.9905946254730225,
|
| 5 |
+
"eval_runtime": 12.3291,
|
| 6 |
+
"eval_samples_per_second": 616.105,
|
| 7 |
+
"eval_steps_per_second": 2.433
|
| 8 |
+
}
|
stress-spelling_error_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 64.63828390062383,
|
| 3 |
+
"eval_combined_score": 64.63828390062383,
|
| 4 |
+
"eval_loss": 0.8153389692306519,
|
| 5 |
+
"eval_runtime": 14.6992,
|
| 6 |
+
"eval_samples_per_second": 621.599,
|
| 7 |
+
"eval_steps_per_second": 2.449
|
| 8 |
+
}
|
stress-word_overlap_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 62.01179820992677,
|
| 3 |
+
"eval_combined_score": 62.01179820992677,
|
| 4 |
+
"eval_loss": 0.8358756303787231,
|
| 5 |
+
"eval_runtime": 15.8566,
|
| 6 |
+
"eval_samples_per_second": 620.058,
|
| 7 |
+
"eval_steps_per_second": 2.46
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"ave/id": 0.6846805289640281,
|
| 3 |
+
"ave/ood": 0.5003333333333333,
|
| 4 |
+
"ave/transfer": 0.5795570006096322,
|
| 5 |
+
"gap/constituent": 0.1836805289640281,
|
| 6 |
+
"gap/lexical_overlap": 0.1846805289640281,
|
| 7 |
+
"gap/ood": 0.1843471956306948,
|
| 8 |
+
"gap/subsequence": 0.1846805289640281,
|
| 9 |
+
"gap/transfer": 0.10512352835439587,
|
| 10 |
+
"mnli-m": 0.6798777381558838,
|
| 11 |
+
"mnli-mm": 0.6894833197721725,
|
| 12 |
+
"ood": 0.5003333333333333,
|
| 13 |
+
"ood/constituent": 0.501,
|
| 14 |
+
"ood/lexical_overlap": 0.5,
|
| 15 |
+
"ood/subsequence": 0.5,
|
| 16 |
+
"transfer": 0.5795570006096322
|
| 17 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_basic_tokenize": true,
|
| 5 |
+
"do_lower_case": true,
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"never_split": null,
|
| 9 |
+
"pad_token": "[PAD]",
|
| 10 |
+
"sep_token": "[SEP]",
|
| 11 |
+
"strip_accents": null,
|
| 12 |
+
"tokenize_chinese_chars": true,
|
| 13 |
+
"tokenizer_class": "BertTokenizer",
|
| 14 |
+
"unk_token": "[UNK]"
|
| 15 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.0,
|
| 3 |
+
"train_loss": 25.68384859034519,
|
| 4 |
+
"train_runtime": 8782.0618,
|
| 5 |
+
"train_samples_per_second": 268.298,
|
| 6 |
+
"train_steps_per_second": 8.384
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,967 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 67.98777381558838,
|
| 3 |
+
"best_model_checkpoint": "outputs/bert-tiny/kw/mnli_21/checkpoint-61360",
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"global_step": 73632,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 1.9864189482833552e-05,
|
| 13 |
+
"loss": 34.5409,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 1.9728378965667103e-05,
|
| 19 |
+
"loss": 32.9539,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.12,
|
| 24 |
+
"learning_rate": 1.9592568448500654e-05,
|
| 25 |
+
"loss": 31.0086,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.16,
|
| 30 |
+
"learning_rate": 1.9456757931334205e-05,
|
| 31 |
+
"loss": 30.5966,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.2,
|
| 36 |
+
"learning_rate": 1.9320947414167756e-05,
|
| 37 |
+
"loss": 30.1725,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.24,
|
| 42 |
+
"learning_rate": 1.9185136897001307e-05,
|
| 43 |
+
"loss": 29.706,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.29,
|
| 48 |
+
"learning_rate": 1.9049326379834858e-05,
|
| 49 |
+
"loss": 29.5048,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.33,
|
| 54 |
+
"learning_rate": 1.8913515862668405e-05,
|
| 55 |
+
"loss": 29.1964,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.37,
|
| 60 |
+
"learning_rate": 1.8777705345501956e-05,
|
| 61 |
+
"loss": 28.8603,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.41,
|
| 66 |
+
"learning_rate": 1.8641894828335507e-05,
|
| 67 |
+
"loss": 28.7968,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.45,
|
| 72 |
+
"learning_rate": 1.8506084311169058e-05,
|
| 73 |
+
"loss": 28.7003,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.49,
|
| 78 |
+
"learning_rate": 1.837027379400261e-05,
|
| 79 |
+
"loss": 28.4954,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.53,
|
| 84 |
+
"learning_rate": 1.823446327683616e-05,
|
| 85 |
+
"loss": 28.24,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.57,
|
| 90 |
+
"learning_rate": 1.809865275966971e-05,
|
| 91 |
+
"loss": 28.1433,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.61,
|
| 96 |
+
"learning_rate": 1.796284224250326e-05,
|
| 97 |
+
"loss": 27.855,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.65,
|
| 102 |
+
"learning_rate": 1.7827031725336812e-05,
|
| 103 |
+
"loss": 27.8772,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.69,
|
| 108 |
+
"learning_rate": 1.7691221208170363e-05,
|
| 109 |
+
"loss": 27.862,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.73,
|
| 114 |
+
"learning_rate": 1.7555410691003914e-05,
|
| 115 |
+
"loss": 27.767,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.77,
|
| 120 |
+
"learning_rate": 1.7419600173837465e-05,
|
| 121 |
+
"loss": 27.6884,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.81,
|
| 126 |
+
"learning_rate": 1.7283789656671015e-05,
|
| 127 |
+
"loss": 27.593,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.86,
|
| 132 |
+
"learning_rate": 1.7147979139504566e-05,
|
| 133 |
+
"loss": 27.4518,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.9,
|
| 138 |
+
"learning_rate": 1.7012168622338114e-05,
|
| 139 |
+
"loss": 27.3718,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.94,
|
| 144 |
+
"learning_rate": 1.6876358105171665e-05,
|
| 145 |
+
"loss": 27.3435,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.98,
|
| 150 |
+
"learning_rate": 1.6740547588005215e-05,
|
| 151 |
+
"loss": 27.3723,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.0,
|
| 156 |
+
"eval_accuracy": 63.87162506367804,
|
| 157 |
+
"eval_combined_score": 63.87162506367804,
|
| 158 |
+
"eval_loss": 0.8185800313949585,
|
| 159 |
+
"eval_runtime": 17.7384,
|
| 160 |
+
"eval_samples_per_second": 553.32,
|
| 161 |
+
"eval_steps_per_second": 2.199,
|
| 162 |
+
"step": 12272
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.02,
|
| 166 |
+
"learning_rate": 1.6604737070838766e-05,
|
| 167 |
+
"loss": 27.0058,
|
| 168 |
+
"step": 12500
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.06,
|
| 172 |
+
"learning_rate": 1.6468926553672317e-05,
|
| 173 |
+
"loss": 26.9092,
|
| 174 |
+
"step": 13000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.1,
|
| 178 |
+
"learning_rate": 1.6333116036505868e-05,
|
| 179 |
+
"loss": 26.8185,
|
| 180 |
+
"step": 13500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.14,
|
| 184 |
+
"learning_rate": 1.619730551933942e-05,
|
| 185 |
+
"loss": 26.6763,
|
| 186 |
+
"step": 14000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.18,
|
| 190 |
+
"learning_rate": 1.606149500217297e-05,
|
| 191 |
+
"loss": 26.7073,
|
| 192 |
+
"step": 14500
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.22,
|
| 196 |
+
"learning_rate": 1.592568448500652e-05,
|
| 197 |
+
"loss": 26.7014,
|
| 198 |
+
"step": 15000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.26,
|
| 202 |
+
"learning_rate": 1.578987396784007e-05,
|
| 203 |
+
"loss": 26.5431,
|
| 204 |
+
"step": 15500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.3,
|
| 208 |
+
"learning_rate": 1.5654063450673622e-05,
|
| 209 |
+
"loss": 26.2987,
|
| 210 |
+
"step": 16000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.34,
|
| 214 |
+
"learning_rate": 1.5518252933507173e-05,
|
| 215 |
+
"loss": 26.4723,
|
| 216 |
+
"step": 16500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.39,
|
| 220 |
+
"learning_rate": 1.5382442416340724e-05,
|
| 221 |
+
"loss": 26.392,
|
| 222 |
+
"step": 17000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.43,
|
| 226 |
+
"learning_rate": 1.5246631899174273e-05,
|
| 227 |
+
"loss": 26.1498,
|
| 228 |
+
"step": 17500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.47,
|
| 232 |
+
"learning_rate": 1.5110821382007822e-05,
|
| 233 |
+
"loss": 26.1262,
|
| 234 |
+
"step": 18000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.51,
|
| 238 |
+
"learning_rate": 1.4975010864841375e-05,
|
| 239 |
+
"loss": 26.1348,
|
| 240 |
+
"step": 18500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.55,
|
| 244 |
+
"learning_rate": 1.4839200347674924e-05,
|
| 245 |
+
"loss": 26.0562,
|
| 246 |
+
"step": 19000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.59,
|
| 250 |
+
"learning_rate": 1.4703389830508477e-05,
|
| 251 |
+
"loss": 26.2436,
|
| 252 |
+
"step": 19500
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.63,
|
| 256 |
+
"learning_rate": 1.4567579313342026e-05,
|
| 257 |
+
"loss": 26.1435,
|
| 258 |
+
"step": 20000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.67,
|
| 262 |
+
"learning_rate": 1.4431768796175577e-05,
|
| 263 |
+
"loss": 26.1062,
|
| 264 |
+
"step": 20500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.71,
|
| 268 |
+
"learning_rate": 1.4295958279009128e-05,
|
| 269 |
+
"loss": 25.9377,
|
| 270 |
+
"step": 21000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.75,
|
| 274 |
+
"learning_rate": 1.4160147761842677e-05,
|
| 275 |
+
"loss": 26.3038,
|
| 276 |
+
"step": 21500
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.79,
|
| 280 |
+
"learning_rate": 1.402433724467623e-05,
|
| 281 |
+
"loss": 26.0764,
|
| 282 |
+
"step": 22000
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.83,
|
| 286 |
+
"learning_rate": 1.3888526727509778e-05,
|
| 287 |
+
"loss": 26.251,
|
| 288 |
+
"step": 22500
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.87,
|
| 292 |
+
"learning_rate": 1.3752716210343331e-05,
|
| 293 |
+
"loss": 25.9488,
|
| 294 |
+
"step": 23000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.91,
|
| 298 |
+
"learning_rate": 1.361690569317688e-05,
|
| 299 |
+
"loss": 25.7248,
|
| 300 |
+
"step": 23500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.96,
|
| 304 |
+
"learning_rate": 1.3481095176010431e-05,
|
| 305 |
+
"loss": 25.5771,
|
| 306 |
+
"step": 24000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 2.0,
|
| 310 |
+
"learning_rate": 1.3345284658843982e-05,
|
| 311 |
+
"loss": 25.9028,
|
| 312 |
+
"step": 24500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.0,
|
| 316 |
+
"eval_accuracy": 65.73611818644932,
|
| 317 |
+
"eval_combined_score": 65.73611818644932,
|
| 318 |
+
"eval_loss": 0.7796743512153625,
|
| 319 |
+
"eval_runtime": 15.909,
|
| 320 |
+
"eval_samples_per_second": 616.945,
|
| 321 |
+
"eval_steps_per_second": 2.451,
|
| 322 |
+
"step": 24544
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.04,
|
| 326 |
+
"learning_rate": 1.3209474141677531e-05,
|
| 327 |
+
"loss": 25.5806,
|
| 328 |
+
"step": 25000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.08,
|
| 332 |
+
"learning_rate": 1.3073663624511084e-05,
|
| 333 |
+
"loss": 25.5793,
|
| 334 |
+
"step": 25500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 2.12,
|
| 338 |
+
"learning_rate": 1.2937853107344633e-05,
|
| 339 |
+
"loss": 25.0838,
|
| 340 |
+
"step": 26000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.16,
|
| 344 |
+
"learning_rate": 1.2802042590178185e-05,
|
| 345 |
+
"loss": 25.2664,
|
| 346 |
+
"step": 26500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.2,
|
| 350 |
+
"learning_rate": 1.2666232073011735e-05,
|
| 351 |
+
"loss": 25.4611,
|
| 352 |
+
"step": 27000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.24,
|
| 356 |
+
"learning_rate": 1.2530421555845285e-05,
|
| 357 |
+
"loss": 25.336,
|
| 358 |
+
"step": 27500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.28,
|
| 362 |
+
"learning_rate": 1.2394611038678836e-05,
|
| 363 |
+
"loss": 25.2676,
|
| 364 |
+
"step": 28000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 2.32,
|
| 368 |
+
"learning_rate": 1.2258800521512385e-05,
|
| 369 |
+
"loss": 25.3788,
|
| 370 |
+
"step": 28500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.36,
|
| 374 |
+
"learning_rate": 1.2122990004345938e-05,
|
| 375 |
+
"loss": 25.0569,
|
| 376 |
+
"step": 29000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.4,
|
| 380 |
+
"learning_rate": 1.1987179487179487e-05,
|
| 381 |
+
"loss": 25.0311,
|
| 382 |
+
"step": 29500
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.44,
|
| 386 |
+
"learning_rate": 1.185136897001304e-05,
|
| 387 |
+
"loss": 24.9824,
|
| 388 |
+
"step": 30000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.49,
|
| 392 |
+
"learning_rate": 1.1715558452846589e-05,
|
| 393 |
+
"loss": 25.2441,
|
| 394 |
+
"step": 30500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.53,
|
| 398 |
+
"learning_rate": 1.157974793568014e-05,
|
| 399 |
+
"loss": 24.9861,
|
| 400 |
+
"step": 31000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.57,
|
| 404 |
+
"learning_rate": 1.144393741851369e-05,
|
| 405 |
+
"loss": 24.7679,
|
| 406 |
+
"step": 31500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.61,
|
| 410 |
+
"learning_rate": 1.130812690134724e-05,
|
| 411 |
+
"loss": 25.271,
|
| 412 |
+
"step": 32000
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 2.65,
|
| 416 |
+
"learning_rate": 1.1172316384180792e-05,
|
| 417 |
+
"loss": 24.9879,
|
| 418 |
+
"step": 32500
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 2.69,
|
| 422 |
+
"learning_rate": 1.1036505867014341e-05,
|
| 423 |
+
"loss": 24.9177,
|
| 424 |
+
"step": 33000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.73,
|
| 428 |
+
"learning_rate": 1.0900695349847894e-05,
|
| 429 |
+
"loss": 25.384,
|
| 430 |
+
"step": 33500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.77,
|
| 434 |
+
"learning_rate": 1.0764884832681443e-05,
|
| 435 |
+
"loss": 25.1108,
|
| 436 |
+
"step": 34000
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.81,
|
| 440 |
+
"learning_rate": 1.0629074315514994e-05,
|
| 441 |
+
"loss": 24.9601,
|
| 442 |
+
"step": 34500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 2.85,
|
| 446 |
+
"learning_rate": 1.0493263798348545e-05,
|
| 447 |
+
"loss": 24.6376,
|
| 448 |
+
"step": 35000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 2.89,
|
| 452 |
+
"learning_rate": 1.0357453281182096e-05,
|
| 453 |
+
"loss": 25.1261,
|
| 454 |
+
"step": 35500
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.93,
|
| 458 |
+
"learning_rate": 1.0221642764015647e-05,
|
| 459 |
+
"loss": 24.8459,
|
| 460 |
+
"step": 36000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.97,
|
| 464 |
+
"learning_rate": 1.0085832246849196e-05,
|
| 465 |
+
"loss": 25.2014,
|
| 466 |
+
"step": 36500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 3.0,
|
| 470 |
+
"eval_accuracy": 66.64289353031076,
|
| 471 |
+
"eval_combined_score": 66.64289353031076,
|
| 472 |
+
"eval_loss": 0.7598663568496704,
|
| 473 |
+
"eval_runtime": 15.7255,
|
| 474 |
+
"eval_samples_per_second": 624.146,
|
| 475 |
+
"eval_steps_per_second": 2.48,
|
| 476 |
+
"step": 36816
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.01,
|
| 480 |
+
"learning_rate": 9.950021729682747e-06,
|
| 481 |
+
"loss": 24.7148,
|
| 482 |
+
"step": 37000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 3.06,
|
| 486 |
+
"learning_rate": 9.814211212516298e-06,
|
| 487 |
+
"loss": 24.4617,
|
| 488 |
+
"step": 37500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.1,
|
| 492 |
+
"learning_rate": 9.678400695349848e-06,
|
| 493 |
+
"loss": 24.7504,
|
| 494 |
+
"step": 38000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.14,
|
| 498 |
+
"learning_rate": 9.5425901781834e-06,
|
| 499 |
+
"loss": 24.45,
|
| 500 |
+
"step": 38500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 3.18,
|
| 504 |
+
"learning_rate": 9.40677966101695e-06,
|
| 505 |
+
"loss": 24.6057,
|
| 506 |
+
"step": 39000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 3.22,
|
| 510 |
+
"learning_rate": 9.2709691438505e-06,
|
| 511 |
+
"loss": 24.3787,
|
| 512 |
+
"step": 39500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.26,
|
| 516 |
+
"learning_rate": 9.13515862668405e-06,
|
| 517 |
+
"loss": 24.2096,
|
| 518 |
+
"step": 40000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 3.3,
|
| 522 |
+
"learning_rate": 8.999348109517601e-06,
|
| 523 |
+
"loss": 24.5361,
|
| 524 |
+
"step": 40500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 3.34,
|
| 528 |
+
"learning_rate": 8.863537592351152e-06,
|
| 529 |
+
"loss": 24.3345,
|
| 530 |
+
"step": 41000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 3.38,
|
| 534 |
+
"learning_rate": 8.727727075184703e-06,
|
| 535 |
+
"loss": 24.364,
|
| 536 |
+
"step": 41500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 3.42,
|
| 540 |
+
"learning_rate": 8.591916558018254e-06,
|
| 541 |
+
"loss": 24.2865,
|
| 542 |
+
"step": 42000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 3.46,
|
| 546 |
+
"learning_rate": 8.456106040851804e-06,
|
| 547 |
+
"loss": 24.5307,
|
| 548 |
+
"step": 42500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 3.5,
|
| 552 |
+
"learning_rate": 8.320295523685355e-06,
|
| 553 |
+
"loss": 24.5127,
|
| 554 |
+
"step": 43000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.54,
|
| 558 |
+
"learning_rate": 8.184485006518904e-06,
|
| 559 |
+
"loss": 24.2719,
|
| 560 |
+
"step": 43500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 3.59,
|
| 564 |
+
"learning_rate": 8.048674489352455e-06,
|
| 565 |
+
"loss": 24.2633,
|
| 566 |
+
"step": 44000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 3.63,
|
| 570 |
+
"learning_rate": 7.912863972186006e-06,
|
| 571 |
+
"loss": 24.1562,
|
| 572 |
+
"step": 44500
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 3.67,
|
| 576 |
+
"learning_rate": 7.777053455019557e-06,
|
| 577 |
+
"loss": 24.1874,
|
| 578 |
+
"step": 45000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 3.71,
|
| 582 |
+
"learning_rate": 7.641242937853108e-06,
|
| 583 |
+
"loss": 24.3553,
|
| 584 |
+
"step": 45500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.75,
|
| 588 |
+
"learning_rate": 7.505432420686659e-06,
|
| 589 |
+
"loss": 24.549,
|
| 590 |
+
"step": 46000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 3.79,
|
| 594 |
+
"learning_rate": 7.369621903520209e-06,
|
| 595 |
+
"loss": 24.538,
|
| 596 |
+
"step": 46500
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 3.83,
|
| 600 |
+
"learning_rate": 7.23381138635376e-06,
|
| 601 |
+
"loss": 24.141,
|
| 602 |
+
"step": 47000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 3.87,
|
| 606 |
+
"learning_rate": 7.0980008691873105e-06,
|
| 607 |
+
"loss": 24.2314,
|
| 608 |
+
"step": 47500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.91,
|
| 612 |
+
"learning_rate": 6.962190352020861e-06,
|
| 613 |
+
"loss": 24.7287,
|
| 614 |
+
"step": 48000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.95,
|
| 618 |
+
"learning_rate": 6.826379834854412e-06,
|
| 619 |
+
"loss": 24.1801,
|
| 620 |
+
"step": 48500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 3.99,
|
| 624 |
+
"learning_rate": 6.690569317687963e-06,
|
| 625 |
+
"loss": 24.3416,
|
| 626 |
+
"step": 49000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 4.0,
|
| 630 |
+
"eval_accuracy": 67.66174223127865,
|
| 631 |
+
"eval_combined_score": 67.66174223127865,
|
| 632 |
+
"eval_loss": 0.7413901090621948,
|
| 633 |
+
"eval_runtime": 15.9956,
|
| 634 |
+
"eval_samples_per_second": 613.605,
|
| 635 |
+
"eval_steps_per_second": 2.438,
|
| 636 |
+
"step": 49088
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 4.03,
|
| 640 |
+
"learning_rate": 6.554758800521513e-06,
|
| 641 |
+
"loss": 23.7173,
|
| 642 |
+
"step": 49500
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 4.07,
|
| 646 |
+
"learning_rate": 6.418948283355063e-06,
|
| 647 |
+
"loss": 24.159,
|
| 648 |
+
"step": 50000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 4.12,
|
| 652 |
+
"learning_rate": 6.283137766188614e-06,
|
| 653 |
+
"loss": 23.9621,
|
| 654 |
+
"step": 50500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 4.16,
|
| 658 |
+
"learning_rate": 6.147327249022165e-06,
|
| 659 |
+
"loss": 24.1423,
|
| 660 |
+
"step": 51000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 4.2,
|
| 664 |
+
"learning_rate": 6.011516731855716e-06,
|
| 665 |
+
"loss": 24.0207,
|
| 666 |
+
"step": 51500
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 4.24,
|
| 670 |
+
"learning_rate": 5.8757062146892665e-06,
|
| 671 |
+
"loss": 24.076,
|
| 672 |
+
"step": 52000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 4.28,
|
| 676 |
+
"learning_rate": 5.739895697522817e-06,
|
| 677 |
+
"loss": 24.0378,
|
| 678 |
+
"step": 52500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 4.32,
|
| 682 |
+
"learning_rate": 5.6040851803563665e-06,
|
| 683 |
+
"loss": 24.0202,
|
| 684 |
+
"step": 53000
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 4.36,
|
| 688 |
+
"learning_rate": 5.468274663189917e-06,
|
| 689 |
+
"loss": 24.0662,
|
| 690 |
+
"step": 53500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 4.4,
|
| 694 |
+
"learning_rate": 5.332464146023468e-06,
|
| 695 |
+
"loss": 24.0337,
|
| 696 |
+
"step": 54000
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 4.44,
|
| 700 |
+
"learning_rate": 5.196653628857019e-06,
|
| 701 |
+
"loss": 23.7662,
|
| 702 |
+
"step": 54500
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 4.48,
|
| 706 |
+
"learning_rate": 5.06084311169057e-06,
|
| 707 |
+
"loss": 23.6851,
|
| 708 |
+
"step": 55000
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 4.52,
|
| 712 |
+
"learning_rate": 4.92503259452412e-06,
|
| 713 |
+
"loss": 23.898,
|
| 714 |
+
"step": 55500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 4.56,
|
| 718 |
+
"learning_rate": 4.789222077357671e-06,
|
| 719 |
+
"loss": 23.7756,
|
| 720 |
+
"step": 56000
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 4.6,
|
| 724 |
+
"learning_rate": 4.653411560191222e-06,
|
| 725 |
+
"loss": 23.9044,
|
| 726 |
+
"step": 56500
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 4.64,
|
| 730 |
+
"learning_rate": 4.5176010430247726e-06,
|
| 731 |
+
"loss": 24.0725,
|
| 732 |
+
"step": 57000
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 4.69,
|
| 736 |
+
"learning_rate": 4.381790525858323e-06,
|
| 737 |
+
"loss": 23.7794,
|
| 738 |
+
"step": 57500
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 4.73,
|
| 742 |
+
"learning_rate": 4.2459800086918734e-06,
|
| 743 |
+
"loss": 24.1247,
|
| 744 |
+
"step": 58000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 4.77,
|
| 748 |
+
"learning_rate": 4.110169491525424e-06,
|
| 749 |
+
"loss": 24.2018,
|
| 750 |
+
"step": 58500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 4.81,
|
| 754 |
+
"learning_rate": 3.974358974358974e-06,
|
| 755 |
+
"loss": 24.1887,
|
| 756 |
+
"step": 59000
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 4.85,
|
| 760 |
+
"learning_rate": 3.838548457192525e-06,
|
| 761 |
+
"loss": 24.0284,
|
| 762 |
+
"step": 59500
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 4.89,
|
| 766 |
+
"learning_rate": 3.702737940026076e-06,
|
| 767 |
+
"loss": 23.9692,
|
| 768 |
+
"step": 60000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 4.93,
|
| 772 |
+
"learning_rate": 3.566927422859627e-06,
|
| 773 |
+
"loss": 23.8436,
|
| 774 |
+
"step": 60500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 4.97,
|
| 778 |
+
"learning_rate": 3.431116905693177e-06,
|
| 779 |
+
"loss": 23.7877,
|
| 780 |
+
"step": 61000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 5.0,
|
| 784 |
+
"eval_accuracy": 67.98777381558838,
|
| 785 |
+
"eval_combined_score": 67.98777381558838,
|
| 786 |
+
"eval_loss": 0.7376300692558289,
|
| 787 |
+
"eval_runtime": 15.7227,
|
| 788 |
+
"eval_samples_per_second": 624.258,
|
| 789 |
+
"eval_steps_per_second": 2.48,
|
| 790 |
+
"step": 61360
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 5.01,
|
| 794 |
+
"learning_rate": 3.2953063885267278e-06,
|
| 795 |
+
"loss": 23.8127,
|
| 796 |
+
"step": 61500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 5.05,
|
| 800 |
+
"learning_rate": 3.1594958713602786e-06,
|
| 801 |
+
"loss": 23.4685,
|
| 802 |
+
"step": 62000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 5.09,
|
| 806 |
+
"learning_rate": 3.0236853541938286e-06,
|
| 807 |
+
"loss": 23.6626,
|
| 808 |
+
"step": 62500
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 5.13,
|
| 812 |
+
"learning_rate": 2.8878748370273795e-06,
|
| 813 |
+
"loss": 23.8414,
|
| 814 |
+
"step": 63000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 5.17,
|
| 818 |
+
"learning_rate": 2.7520643198609304e-06,
|
| 819 |
+
"loss": 23.6328,
|
| 820 |
+
"step": 63500
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 5.22,
|
| 824 |
+
"learning_rate": 2.6162538026944812e-06,
|
| 825 |
+
"loss": 23.9316,
|
| 826 |
+
"step": 64000
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 5.26,
|
| 830 |
+
"learning_rate": 2.4804432855280312e-06,
|
| 831 |
+
"loss": 24.1051,
|
| 832 |
+
"step": 64500
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 5.3,
|
| 836 |
+
"learning_rate": 2.344632768361582e-06,
|
| 837 |
+
"loss": 23.7414,
|
| 838 |
+
"step": 65000
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 5.34,
|
| 842 |
+
"learning_rate": 2.2088222511951325e-06,
|
| 843 |
+
"loss": 23.8935,
|
| 844 |
+
"step": 65500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 5.38,
|
| 848 |
+
"learning_rate": 2.0730117340286834e-06,
|
| 849 |
+
"loss": 23.6011,
|
| 850 |
+
"step": 66000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 5.42,
|
| 854 |
+
"learning_rate": 1.937201216862234e-06,
|
| 855 |
+
"loss": 23.6583,
|
| 856 |
+
"step": 66500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 5.46,
|
| 860 |
+
"learning_rate": 1.8013906996957845e-06,
|
| 861 |
+
"loss": 23.4516,
|
| 862 |
+
"step": 67000
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 5.5,
|
| 866 |
+
"learning_rate": 1.6655801825293353e-06,
|
| 867 |
+
"loss": 23.9621,
|
| 868 |
+
"step": 67500
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.54,
|
| 872 |
+
"learning_rate": 1.5297696653628858e-06,
|
| 873 |
+
"loss": 23.5286,
|
| 874 |
+
"step": 68000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 5.58,
|
| 878 |
+
"learning_rate": 1.3939591481964364e-06,
|
| 879 |
+
"loss": 23.7543,
|
| 880 |
+
"step": 68500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 5.62,
|
| 884 |
+
"learning_rate": 1.258148631029987e-06,
|
| 885 |
+
"loss": 23.961,
|
| 886 |
+
"step": 69000
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 5.66,
|
| 890 |
+
"learning_rate": 1.1223381138635377e-06,
|
| 891 |
+
"loss": 23.4363,
|
| 892 |
+
"step": 69500
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 5.7,
|
| 896 |
+
"learning_rate": 9.865275966970884e-07,
|
| 897 |
+
"loss": 23.5779,
|
| 898 |
+
"step": 70000
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 5.74,
|
| 902 |
+
"learning_rate": 8.507170795306389e-07,
|
| 903 |
+
"loss": 23.9706,
|
| 904 |
+
"step": 70500
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 5.79,
|
| 908 |
+
"learning_rate": 7.149065623641896e-07,
|
| 909 |
+
"loss": 23.6523,
|
| 910 |
+
"step": 71000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 5.83,
|
| 914 |
+
"learning_rate": 5.790960451977402e-07,
|
| 915 |
+
"loss": 23.5014,
|
| 916 |
+
"step": 71500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 5.87,
|
| 920 |
+
"learning_rate": 4.432855280312908e-07,
|
| 921 |
+
"loss": 23.6879,
|
| 922 |
+
"step": 72000
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 5.91,
|
| 926 |
+
"learning_rate": 3.074750108648414e-07,
|
| 927 |
+
"loss": 65.8195,
|
| 928 |
+
"step": 72500
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 5.95,
|
| 932 |
+
"learning_rate": 1.7166449369839202e-07,
|
| 933 |
+
"loss": 23.5878,
|
| 934 |
+
"step": 73000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 5.99,
|
| 938 |
+
"learning_rate": 3.585397653194264e-08,
|
| 939 |
+
"loss": 23.5938,
|
| 940 |
+
"step": 73500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 6.0,
|
| 944 |
+
"eval_accuracy": 67.77381558838512,
|
| 945 |
+
"eval_combined_score": 67.77381558838512,
|
| 946 |
+
"eval_loss": 0.7389720678329468,
|
| 947 |
+
"eval_runtime": 15.833,
|
| 948 |
+
"eval_samples_per_second": 619.906,
|
| 949 |
+
"eval_steps_per_second": 2.463,
|
| 950 |
+
"step": 73632
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 6.0,
|
| 954 |
+
"step": 73632,
|
| 955 |
+
"total_flos": 0.0,
|
| 956 |
+
"train_loss": 25.68384859034519,
|
| 957 |
+
"train_runtime": 8782.0618,
|
| 958 |
+
"train_samples_per_second": 268.298,
|
| 959 |
+
"train_steps_per_second": 8.384
|
| 960 |
+
}
|
| 961 |
+
],
|
| 962 |
+
"max_steps": 73632,
|
| 963 |
+
"num_train_epochs": 6,
|
| 964 |
+
"total_flos": 0.0,
|
| 965 |
+
"trial_name": null,
|
| 966 |
+
"trial_params": null
|
| 967 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e969d308a6224c6afd8946c2f2c5e1ffcb51318a9a834f0a0c2ae4689bd018fd
|
| 3 |
+
size 4283
|
transfer-qnli_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 51.41863444993593,
|
| 3 |
+
"eval_combined_score": 51.41863444993593,
|
| 4 |
+
"eval_loss": 1.8891557455062866,
|
| 5 |
+
"eval_runtime": 7.6329,
|
| 6 |
+
"eval_samples_per_second": 715.722,
|
| 7 |
+
"eval_steps_per_second": 2.882
|
| 8 |
+
}
|
transfer-snli_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 57.955700060963224,
|
| 3 |
+
"eval_combined_score": 57.955700060963224,
|
| 4 |
+
"eval_loss": 0.8794325590133667,
|
| 5 |
+
"eval_runtime": 15.2962,
|
| 6 |
+
"eval_samples_per_second": 643.427,
|
| 7 |
+
"eval_steps_per_second": 2.55
|
| 8 |
+
}
|
transfer_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.5795570006096322,
|
| 3 |
+
"eval_combined_score": 0.5795570006096322,
|
| 4 |
+
"eval_loss": 0.8794325590133667,
|
| 5 |
+
"eval_runtime": 3.9813,
|
| 6 |
+
"eval_samples_per_second": 2472.044,
|
| 7 |
+
"eval_steps_per_second": 9.796
|
| 8 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|