Initial commit from mrm8488
Browse files- README.md +147 -0
- added_tokens.json +1 -0
- all_results.json +14 -0
- config.json +46 -0
- eval_results.json +9 -0
- pytorch_model.bin +3 -0
- runs/events.out.tfevents.1637496578.98522726b8f6.383.0 +3 -0
- runs/events.out.tfevents.1637513130.98522726b8f6.383.2 +3 -0
- special_tokens_map.json +1 -0
- spm.model +3 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1120 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,147 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: mit
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- glue
|
| 9 |
+
metrics:
|
| 10 |
+
- accuracy
|
| 11 |
+
model-index:
|
| 12 |
+
- name: ds_results
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Text Classification
|
| 16 |
+
type: text-classification
|
| 17 |
+
dataset:
|
| 18 |
+
name: GLUE MNLI
|
| 19 |
+
type: glue
|
| 20 |
+
args: mnli
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Accuracy
|
| 23 |
+
type: accuracy
|
| 24 |
+
value: 0.874593165174939
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 28 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 29 |
+
|
| 30 |
+
# ds_results
|
| 31 |
+
|
| 32 |
+
This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the GLUE MNLI dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 0.4985
|
| 35 |
+
- Accuracy: 0.8746
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 3e-05
|
| 55 |
+
- train_batch_size: 16
|
| 56 |
+
- eval_batch_size: 8
|
| 57 |
+
- seed: 42
|
| 58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 59 |
+
- lr_scheduler_type: linear
|
| 60 |
+
- lr_scheduler_warmup_steps: 1000
|
| 61 |
+
- num_epochs: 3.0
|
| 62 |
+
|
| 63 |
+
### Training results
|
| 64 |
+
|
| 65 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 66 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 67 |
+
| 0.7773 | 0.04 | 1000 | 0.5241 | 0.7984 |
|
| 68 |
+
| 0.546 | 0.08 | 2000 | 0.4629 | 0.8194 |
|
| 69 |
+
| 0.5032 | 0.12 | 3000 | 0.4704 | 0.8274 |
|
| 70 |
+
| 0.4711 | 0.16 | 4000 | 0.4383 | 0.8355 |
|
| 71 |
+
| 0.473 | 0.2 | 5000 | 0.4652 | 0.8305 |
|
| 72 |
+
| 0.4619 | 0.24 | 6000 | 0.4234 | 0.8386 |
|
| 73 |
+
| 0.4542 | 0.29 | 7000 | 0.4825 | 0.8349 |
|
| 74 |
+
| 0.4468 | 0.33 | 8000 | 0.3985 | 0.8513 |
|
| 75 |
+
| 0.4288 | 0.37 | 9000 | 0.4084 | 0.8493 |
|
| 76 |
+
| 0.4354 | 0.41 | 10000 | 0.3850 | 0.8533 |
|
| 77 |
+
| 0.423 | 0.45 | 11000 | 0.3855 | 0.8509 |
|
| 78 |
+
| 0.4167 | 0.49 | 12000 | 0.4122 | 0.8513 |
|
| 79 |
+
| 0.4129 | 0.53 | 13000 | 0.4009 | 0.8550 |
|
| 80 |
+
| 0.4135 | 0.57 | 14000 | 0.4136 | 0.8544 |
|
| 81 |
+
| 0.4074 | 0.61 | 15000 | 0.3869 | 0.8595 |
|
| 82 |
+
| 0.415 | 0.65 | 16000 | 0.3911 | 0.8517 |
|
| 83 |
+
| 0.4095 | 0.69 | 17000 | 0.3880 | 0.8593 |
|
| 84 |
+
| 0.4001 | 0.73 | 18000 | 0.3907 | 0.8587 |
|
| 85 |
+
| 0.4069 | 0.77 | 19000 | 0.3686 | 0.8630 |
|
| 86 |
+
| 0.3927 | 0.81 | 20000 | 0.4008 | 0.8593 |
|
| 87 |
+
| 0.3958 | 0.86 | 21000 | 0.3716 | 0.8639 |
|
| 88 |
+
| 0.4016 | 0.9 | 22000 | 0.3594 | 0.8679 |
|
| 89 |
+
| 0.3945 | 0.94 | 23000 | 0.3595 | 0.8679 |
|
| 90 |
+
| 0.3932 | 0.98 | 24000 | 0.3577 | 0.8645 |
|
| 91 |
+
| 0.345 | 1.02 | 25000 | 0.4080 | 0.8699 |
|
| 92 |
+
| 0.2885 | 1.06 | 26000 | 0.3919 | 0.8674 |
|
| 93 |
+
| 0.2858 | 1.1 | 27000 | 0.4346 | 0.8651 |
|
| 94 |
+
| 0.2872 | 1.14 | 28000 | 0.4105 | 0.8674 |
|
| 95 |
+
| 0.3002 | 1.18 | 29000 | 0.4133 | 0.8708 |
|
| 96 |
+
| 0.2954 | 1.22 | 30000 | 0.4062 | 0.8667 |
|
| 97 |
+
| 0.2912 | 1.26 | 31000 | 0.3972 | 0.8708 |
|
| 98 |
+
| 0.2958 | 1.3 | 32000 | 0.3713 | 0.8732 |
|
| 99 |
+
| 0.293 | 1.34 | 33000 | 0.3717 | 0.8715 |
|
| 100 |
+
| 0.3001 | 1.39 | 34000 | 0.3826 | 0.8716 |
|
| 101 |
+
| 0.2864 | 1.43 | 35000 | 0.4155 | 0.8694 |
|
| 102 |
+
| 0.2827 | 1.47 | 36000 | 0.4224 | 0.8666 |
|
| 103 |
+
| 0.2836 | 1.51 | 37000 | 0.3832 | 0.8744 |
|
| 104 |
+
| 0.2844 | 1.55 | 38000 | 0.4179 | 0.8699 |
|
| 105 |
+
| 0.2866 | 1.59 | 39000 | 0.3969 | 0.8681 |
|
| 106 |
+
| 0.2883 | 1.63 | 40000 | 0.4000 | 0.8683 |
|
| 107 |
+
| 0.2832 | 1.67 | 41000 | 0.3853 | 0.8688 |
|
| 108 |
+
| 0.2876 | 1.71 | 42000 | 0.3924 | 0.8677 |
|
| 109 |
+
| 0.2855 | 1.75 | 43000 | 0.4177 | 0.8719 |
|
| 110 |
+
| 0.2845 | 1.79 | 44000 | 0.3877 | 0.8724 |
|
| 111 |
+
| 0.2882 | 1.83 | 45000 | 0.3961 | 0.8713 |
|
| 112 |
+
| 0.2773 | 1.87 | 46000 | 0.3791 | 0.8740 |
|
| 113 |
+
| 0.2767 | 1.91 | 47000 | 0.3877 | 0.8779 |
|
| 114 |
+
| 0.2772 | 1.96 | 48000 | 0.4022 | 0.8690 |
|
| 115 |
+
| 0.2816 | 2.0 | 49000 | 0.3837 | 0.8732 |
|
| 116 |
+
| 0.2068 | 2.04 | 50000 | 0.4644 | 0.8720 |
|
| 117 |
+
| 0.1914 | 2.08 | 51000 | 0.4919 | 0.8744 |
|
| 118 |
+
| 0.2 | 2.12 | 52000 | 0.4870 | 0.8702 |
|
| 119 |
+
| 0.1904 | 2.16 | 53000 | 0.5038 | 0.8737 |
|
| 120 |
+
| 0.1915 | 2.2 | 54000 | 0.5232 | 0.8711 |
|
| 121 |
+
| 0.1956 | 2.24 | 55000 | 0.5192 | 0.8747 |
|
| 122 |
+
| 0.1911 | 2.28 | 56000 | 0.5215 | 0.8761 |
|
| 123 |
+
| 0.2053 | 2.32 | 57000 | 0.4604 | 0.8738 |
|
| 124 |
+
| 0.2008 | 2.36 | 58000 | 0.5162 | 0.8715 |
|
| 125 |
+
| 0.1971 | 2.4 | 59000 | 0.4886 | 0.8754 |
|
| 126 |
+
| 0.192 | 2.44 | 60000 | 0.4921 | 0.8725 |
|
| 127 |
+
| 0.1937 | 2.49 | 61000 | 0.4917 | 0.8763 |
|
| 128 |
+
| 0.1931 | 2.53 | 62000 | 0.4789 | 0.8778 |
|
| 129 |
+
| 0.1964 | 2.57 | 63000 | 0.4997 | 0.8721 |
|
| 130 |
+
| 0.2008 | 2.61 | 64000 | 0.4748 | 0.8756 |
|
| 131 |
+
| 0.1962 | 2.65 | 65000 | 0.4840 | 0.8764 |
|
| 132 |
+
| 0.2029 | 2.69 | 66000 | 0.4889 | 0.8767 |
|
| 133 |
+
| 0.1927 | 2.73 | 67000 | 0.4820 | 0.8758 |
|
| 134 |
+
| 0.1926 | 2.77 | 68000 | 0.4857 | 0.8762 |
|
| 135 |
+
| 0.1919 | 2.81 | 69000 | 0.4836 | 0.8749 |
|
| 136 |
+
| 0.1911 | 2.85 | 70000 | 0.4859 | 0.8742 |
|
| 137 |
+
| 0.1897 | 2.89 | 71000 | 0.4853 | 0.8766 |
|
| 138 |
+
| 0.186 | 2.93 | 72000 | 0.4946 | 0.8768 |
|
| 139 |
+
| 0.2011 | 2.97 | 73000 | 0.4851 | 0.8767 |
|
| 140 |
+
|
| 141 |
+
|
| 142 |
+
### Framework versions
|
| 143 |
+
|
| 144 |
+
- Transformers 4.13.0.dev0
|
| 145 |
+
- Pytorch 1.10.0+cu111
|
| 146 |
+
- Datasets 1.15.1
|
| 147 |
+
- Tokenizers 0.10.3
|
added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"[MASK]": 128000}
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.874593165174939,
|
| 4 |
+
"eval_loss": 0.49852925539016724,
|
| 5 |
+
"eval_runtime": 42.4774,
|
| 6 |
+
"eval_samples": 9832,
|
| 7 |
+
"eval_samples_per_second": 231.464,
|
| 8 |
+
"eval_steps_per_second": 28.933,
|
| 9 |
+
"train_loss": 0.3087343405972663,
|
| 10 |
+
"train_runtime": 16507.3954,
|
| 11 |
+
"train_samples": 392702,
|
| 12 |
+
"train_samples_per_second": 71.368,
|
| 13 |
+
"train_steps_per_second": 4.461
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-v3-small",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"finetuning_task": "mnli",
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "entailment",
|
| 13 |
+
"1": "neutral",
|
| 14 |
+
"2": "contradiction"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"contradiction": 2,
|
| 20 |
+
"entailment": 0,
|
| 21 |
+
"neutral": 1
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-07,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"max_relative_positions": -1,
|
| 26 |
+
"model_type": "deberta-v2",
|
| 27 |
+
"norm_rel_ebd": "layer_norm",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 6,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"pooler_dropout": 0,
|
| 32 |
+
"pooler_hidden_act": "gelu",
|
| 33 |
+
"pooler_hidden_size": 768,
|
| 34 |
+
"pos_att_type": [
|
| 35 |
+
"p2c",
|
| 36 |
+
"c2p"
|
| 37 |
+
],
|
| 38 |
+
"position_biased_input": false,
|
| 39 |
+
"position_buckets": 256,
|
| 40 |
+
"relative_attention": true,
|
| 41 |
+
"share_att_key": true,
|
| 42 |
+
"torch_dtype": "float32",
|
| 43 |
+
"transformers_version": "4.13.0.dev0",
|
| 44 |
+
"type_vocab_size": 0,
|
| 45 |
+
"vocab_size": 128100
|
| 46 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.874593165174939,
|
| 4 |
+
"eval_loss": 0.49852925539016724,
|
| 5 |
+
"eval_runtime": 42.4774,
|
| 6 |
+
"eval_samples": 9832,
|
| 7 |
+
"eval_samples_per_second": 231.464,
|
| 8 |
+
"eval_steps_per_second": 28.933
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:533b3363d914bc2b61131e395088a9e89938156bc0d8841202dc3bd78cfbbfb2
|
| 3 |
+
size 567635947
|
runs/events.out.tfevents.1637496578.98522726b8f6.383.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10400eb0d7dd0a39f41242ecd14a3dcf4338a5c7b814b3137295d75baf53b7b5
|
| 3 |
+
size 39390
|
runs/events.out.tfevents.1637513130.98522726b8f6.383.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b1f2030960adbe25d647762fdafafe34ed27e335a846c2d38ae4d0bd19b5a37
|
| 3 |
+
size 698
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": false, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "split_by_punct": false, "sp_model_kwargs": {}, "vocab_type": "spm", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "microsoft/deberta-v3-small", "tokenizer_class": "DebertaV2Tokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.3087343405972663,
|
| 4 |
+
"train_runtime": 16507.3954,
|
| 5 |
+
"train_samples": 392702,
|
| 6 |
+
"train_samples_per_second": 71.368,
|
| 7 |
+
"train_steps_per_second": 4.461
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1120 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 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": 3e-05,
|
| 13 |
+
"loss": 0.7773,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"eval_accuracy": 0.7983698420784513,
|
| 19 |
+
"eval_loss": 0.5241270065307617,
|
| 20 |
+
"eval_runtime": 42.0562,
|
| 21 |
+
"eval_samples_per_second": 233.378,
|
| 22 |
+
"eval_steps_per_second": 29.175,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08,
|
| 27 |
+
"learning_rate": 2.9586958916180196e-05,
|
| 28 |
+
"loss": 0.546,
|
| 29 |
+
"step": 2000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.08,
|
| 33 |
+
"eval_accuracy": 0.8193581253183903,
|
| 34 |
+
"eval_loss": 0.46290943026542664,
|
| 35 |
+
"eval_runtime": 42.1932,
|
| 36 |
+
"eval_samples_per_second": 232.621,
|
| 37 |
+
"eval_steps_per_second": 29.081,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"learning_rate": 2.9173917832360394e-05,
|
| 43 |
+
"loss": 0.5032,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"eval_accuracy": 0.8274070300560367,
|
| 49 |
+
"eval_loss": 0.4703749418258667,
|
| 50 |
+
"eval_runtime": 42.1306,
|
| 51 |
+
"eval_samples_per_second": 232.966,
|
| 52 |
+
"eval_steps_per_second": 29.124,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.16,
|
| 57 |
+
"learning_rate": 2.876087674854059e-05,
|
| 58 |
+
"loss": 0.4711,
|
| 59 |
+
"step": 4000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16,
|
| 63 |
+
"eval_accuracy": 0.8354559347936832,
|
| 64 |
+
"eval_loss": 0.4382944405078888,
|
| 65 |
+
"eval_runtime": 41.9141,
|
| 66 |
+
"eval_samples_per_second": 234.169,
|
| 67 |
+
"eval_steps_per_second": 29.274,
|
| 68 |
+
"step": 4000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.2,
|
| 72 |
+
"learning_rate": 2.8347835664720783e-05,
|
| 73 |
+
"loss": 0.473,
|
| 74 |
+
"step": 5000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.2,
|
| 78 |
+
"eval_accuracy": 0.8304635761589404,
|
| 79 |
+
"eval_loss": 0.4652259051799774,
|
| 80 |
+
"eval_runtime": 42.5911,
|
| 81 |
+
"eval_samples_per_second": 230.447,
|
| 82 |
+
"eval_steps_per_second": 28.809,
|
| 83 |
+
"step": 5000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.24,
|
| 87 |
+
"learning_rate": 2.793479458090098e-05,
|
| 88 |
+
"loss": 0.4619,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.24,
|
| 93 |
+
"eval_accuracy": 0.8386143657666837,
|
| 94 |
+
"eval_loss": 0.42338472604751587,
|
| 95 |
+
"eval_runtime": 42.1336,
|
| 96 |
+
"eval_samples_per_second": 232.949,
|
| 97 |
+
"eval_steps_per_second": 29.122,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.29,
|
| 102 |
+
"learning_rate": 2.7521753497081176e-05,
|
| 103 |
+
"loss": 0.4542,
|
| 104 |
+
"step": 7000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.29,
|
| 108 |
+
"eval_accuracy": 0.8349465104431992,
|
| 109 |
+
"eval_loss": 0.48245278000831604,
|
| 110 |
+
"eval_runtime": 42.4819,
|
| 111 |
+
"eval_samples_per_second": 231.04,
|
| 112 |
+
"eval_steps_per_second": 28.883,
|
| 113 |
+
"step": 7000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.33,
|
| 117 |
+
"learning_rate": 2.7108712413261374e-05,
|
| 118 |
+
"loss": 0.4468,
|
| 119 |
+
"step": 8000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.33,
|
| 123 |
+
"eval_accuracy": 0.8513499745287825,
|
| 124 |
+
"eval_loss": 0.39848873019218445,
|
| 125 |
+
"eval_runtime": 42.353,
|
| 126 |
+
"eval_samples_per_second": 231.743,
|
| 127 |
+
"eval_steps_per_second": 28.971,
|
| 128 |
+
"step": 8000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.37,
|
| 132 |
+
"learning_rate": 2.669567132944157e-05,
|
| 133 |
+
"loss": 0.4288,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.37,
|
| 138 |
+
"eval_accuracy": 0.8493122771268466,
|
| 139 |
+
"eval_loss": 0.40836581587791443,
|
| 140 |
+
"eval_runtime": 42.666,
|
| 141 |
+
"eval_samples_per_second": 230.042,
|
| 142 |
+
"eval_steps_per_second": 28.758,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.41,
|
| 147 |
+
"learning_rate": 2.6282630245621764e-05,
|
| 148 |
+
"loss": 0.4354,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.41,
|
| 153 |
+
"eval_accuracy": 0.8532857870606215,
|
| 154 |
+
"eval_loss": 0.3850448429584503,
|
| 155 |
+
"eval_runtime": 42.6387,
|
| 156 |
+
"eval_samples_per_second": 230.19,
|
| 157 |
+
"eval_steps_per_second": 28.777,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.45,
|
| 162 |
+
"learning_rate": 2.5869589161801962e-05,
|
| 163 |
+
"loss": 0.423,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.45,
|
| 168 |
+
"eval_accuracy": 0.8509424350483953,
|
| 169 |
+
"eval_loss": 0.3855280578136444,
|
| 170 |
+
"eval_runtime": 42.4952,
|
| 171 |
+
"eval_samples_per_second": 230.967,
|
| 172 |
+
"eval_steps_per_second": 28.874,
|
| 173 |
+
"step": 11000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.49,
|
| 177 |
+
"learning_rate": 2.5456548077982157e-05,
|
| 178 |
+
"loss": 0.4167,
|
| 179 |
+
"step": 12000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.49,
|
| 183 |
+
"eval_accuracy": 0.8513499745287825,
|
| 184 |
+
"eval_loss": 0.41219788789749146,
|
| 185 |
+
"eval_runtime": 42.4026,
|
| 186 |
+
"eval_samples_per_second": 231.472,
|
| 187 |
+
"eval_steps_per_second": 28.937,
|
| 188 |
+
"step": 12000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.53,
|
| 192 |
+
"learning_rate": 2.504350699416235e-05,
|
| 193 |
+
"loss": 0.4129,
|
| 194 |
+
"step": 13000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.53,
|
| 198 |
+
"eval_accuracy": 0.8550178298522669,
|
| 199 |
+
"eval_loss": 0.40088921785354614,
|
| 200 |
+
"eval_runtime": 42.5886,
|
| 201 |
+
"eval_samples_per_second": 230.461,
|
| 202 |
+
"eval_steps_per_second": 28.811,
|
| 203 |
+
"step": 13000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.57,
|
| 207 |
+
"learning_rate": 2.463046591034255e-05,
|
| 208 |
+
"loss": 0.4135,
|
| 209 |
+
"step": 14000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.57,
|
| 213 |
+
"eval_accuracy": 0.8544065206316862,
|
| 214 |
+
"eval_loss": 0.4136492609977722,
|
| 215 |
+
"eval_runtime": 42.4803,
|
| 216 |
+
"eval_samples_per_second": 231.048,
|
| 217 |
+
"eval_steps_per_second": 28.884,
|
| 218 |
+
"step": 14000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.61,
|
| 222 |
+
"learning_rate": 2.4217424826522744e-05,
|
| 223 |
+
"loss": 0.4074,
|
| 224 |
+
"step": 15000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.61,
|
| 228 |
+
"eval_accuracy": 0.8595007641365258,
|
| 229 |
+
"eval_loss": 0.38692933320999146,
|
| 230 |
+
"eval_runtime": 42.276,
|
| 231 |
+
"eval_samples_per_second": 232.165,
|
| 232 |
+
"eval_steps_per_second": 29.024,
|
| 233 |
+
"step": 15000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.65,
|
| 237 |
+
"learning_rate": 2.3804383742702942e-05,
|
| 238 |
+
"loss": 0.415,
|
| 239 |
+
"step": 16000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.65,
|
| 243 |
+
"eval_accuracy": 0.8516556291390729,
|
| 244 |
+
"eval_loss": 0.39110422134399414,
|
| 245 |
+
"eval_runtime": 42.4226,
|
| 246 |
+
"eval_samples_per_second": 231.363,
|
| 247 |
+
"eval_steps_per_second": 28.923,
|
| 248 |
+
"step": 16000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.69,
|
| 252 |
+
"learning_rate": 2.3391342658883137e-05,
|
| 253 |
+
"loss": 0.4095,
|
| 254 |
+
"step": 17000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.69,
|
| 258 |
+
"eval_accuracy": 0.8592969943963321,
|
| 259 |
+
"eval_loss": 0.38802793622016907,
|
| 260 |
+
"eval_runtime": 42.6242,
|
| 261 |
+
"eval_samples_per_second": 230.268,
|
| 262 |
+
"eval_steps_per_second": 28.786,
|
| 263 |
+
"step": 17000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.73,
|
| 267 |
+
"learning_rate": 2.2978301575063332e-05,
|
| 268 |
+
"loss": 0.4001,
|
| 269 |
+
"step": 18000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.73,
|
| 273 |
+
"eval_accuracy": 0.8586856851757514,
|
| 274 |
+
"eval_loss": 0.3907186985015869,
|
| 275 |
+
"eval_runtime": 42.4988,
|
| 276 |
+
"eval_samples_per_second": 230.948,
|
| 277 |
+
"eval_steps_per_second": 28.871,
|
| 278 |
+
"step": 18000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.77,
|
| 282 |
+
"learning_rate": 2.256526049124353e-05,
|
| 283 |
+
"loss": 0.4069,
|
| 284 |
+
"step": 19000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.77,
|
| 288 |
+
"eval_accuracy": 0.8629648497198166,
|
| 289 |
+
"eval_loss": 0.3686215281486511,
|
| 290 |
+
"eval_runtime": 42.3356,
|
| 291 |
+
"eval_samples_per_second": 231.838,
|
| 292 |
+
"eval_steps_per_second": 28.983,
|
| 293 |
+
"step": 19000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.81,
|
| 297 |
+
"learning_rate": 2.2152219407423725e-05,
|
| 298 |
+
"loss": 0.3927,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.81,
|
| 303 |
+
"eval_accuracy": 0.8592969943963321,
|
| 304 |
+
"eval_loss": 0.4008384943008423,
|
| 305 |
+
"eval_runtime": 42.4047,
|
| 306 |
+
"eval_samples_per_second": 231.46,
|
| 307 |
+
"eval_steps_per_second": 28.935,
|
| 308 |
+
"step": 20000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.86,
|
| 312 |
+
"learning_rate": 2.173917832360392e-05,
|
| 313 |
+
"loss": 0.3958,
|
| 314 |
+
"step": 21000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.86,
|
| 318 |
+
"eval_accuracy": 0.8638818135506877,
|
| 319 |
+
"eval_loss": 0.37160658836364746,
|
| 320 |
+
"eval_runtime": 42.5674,
|
| 321 |
+
"eval_samples_per_second": 230.575,
|
| 322 |
+
"eval_steps_per_second": 28.825,
|
| 323 |
+
"step": 21000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.9,
|
| 327 |
+
"learning_rate": 2.1326137239784118e-05,
|
| 328 |
+
"loss": 0.4016,
|
| 329 |
+
"step": 22000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.9,
|
| 333 |
+
"eval_accuracy": 0.8678553234844626,
|
| 334 |
+
"eval_loss": 0.35941794514656067,
|
| 335 |
+
"eval_runtime": 42.472,
|
| 336 |
+
"eval_samples_per_second": 231.093,
|
| 337 |
+
"eval_steps_per_second": 28.89,
|
| 338 |
+
"step": 22000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.94,
|
| 342 |
+
"learning_rate": 2.0913096155964313e-05,
|
| 343 |
+
"loss": 0.3945,
|
| 344 |
+
"step": 23000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.94,
|
| 348 |
+
"eval_accuracy": 0.8678553234844626,
|
| 349 |
+
"eval_loss": 0.359527587890625,
|
| 350 |
+
"eval_runtime": 42.614,
|
| 351 |
+
"eval_samples_per_second": 230.324,
|
| 352 |
+
"eval_steps_per_second": 28.793,
|
| 353 |
+
"step": 23000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.98,
|
| 357 |
+
"learning_rate": 2.050005507214451e-05,
|
| 358 |
+
"loss": 0.3932,
|
| 359 |
+
"step": 24000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.98,
|
| 363 |
+
"eval_accuracy": 0.8644931227712684,
|
| 364 |
+
"eval_loss": 0.35774311423301697,
|
| 365 |
+
"eval_runtime": 42.4615,
|
| 366 |
+
"eval_samples_per_second": 231.151,
|
| 367 |
+
"eval_steps_per_second": 28.897,
|
| 368 |
+
"step": 24000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.02,
|
| 372 |
+
"learning_rate": 2.0087013988324705e-05,
|
| 373 |
+
"loss": 0.345,
|
| 374 |
+
"step": 25000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.02,
|
| 378 |
+
"eval_accuracy": 0.8698930208863984,
|
| 379 |
+
"eval_loss": 0.40802302956581116,
|
| 380 |
+
"eval_runtime": 42.6011,
|
| 381 |
+
"eval_samples_per_second": 230.393,
|
| 382 |
+
"eval_steps_per_second": 28.802,
|
| 383 |
+
"step": 25000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.06,
|
| 387 |
+
"learning_rate": 1.96739729045049e-05,
|
| 388 |
+
"loss": 0.2885,
|
| 389 |
+
"step": 26000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.06,
|
| 393 |
+
"eval_accuracy": 0.8674477840040754,
|
| 394 |
+
"eval_loss": 0.39192140102386475,
|
| 395 |
+
"eval_runtime": 42.4976,
|
| 396 |
+
"eval_samples_per_second": 230.954,
|
| 397 |
+
"eval_steps_per_second": 28.872,
|
| 398 |
+
"step": 26000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.1,
|
| 402 |
+
"learning_rate": 1.92609318206851e-05,
|
| 403 |
+
"loss": 0.2858,
|
| 404 |
+
"step": 27000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.1,
|
| 408 |
+
"eval_accuracy": 0.8651044319918492,
|
| 409 |
+
"eval_loss": 0.4346281588077545,
|
| 410 |
+
"eval_runtime": 42.3673,
|
| 411 |
+
"eval_samples_per_second": 231.664,
|
| 412 |
+
"eval_steps_per_second": 28.961,
|
| 413 |
+
"step": 27000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.14,
|
| 417 |
+
"learning_rate": 1.8847890736865293e-05,
|
| 418 |
+
"loss": 0.2872,
|
| 419 |
+
"step": 28000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.14,
|
| 423 |
+
"eval_accuracy": 0.8674477840040754,
|
| 424 |
+
"eval_loss": 0.41050681471824646,
|
| 425 |
+
"eval_runtime": 42.4953,
|
| 426 |
+
"eval_samples_per_second": 230.967,
|
| 427 |
+
"eval_steps_per_second": 28.874,
|
| 428 |
+
"step": 28000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.18,
|
| 432 |
+
"learning_rate": 1.8434849653045488e-05,
|
| 433 |
+
"loss": 0.3002,
|
| 434 |
+
"step": 29000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.18,
|
| 438 |
+
"eval_accuracy": 0.8708099847172694,
|
| 439 |
+
"eval_loss": 0.4133119583129883,
|
| 440 |
+
"eval_runtime": 42.4588,
|
| 441 |
+
"eval_samples_per_second": 231.165,
|
| 442 |
+
"eval_steps_per_second": 28.899,
|
| 443 |
+
"step": 29000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.22,
|
| 447 |
+
"learning_rate": 1.8021808569225686e-05,
|
| 448 |
+
"loss": 0.2954,
|
| 449 |
+
"step": 30000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.22,
|
| 453 |
+
"eval_accuracy": 0.8667345899133979,
|
| 454 |
+
"eval_loss": 0.406183123588562,
|
| 455 |
+
"eval_runtime": 42.2388,
|
| 456 |
+
"eval_samples_per_second": 232.369,
|
| 457 |
+
"eval_steps_per_second": 29.049,
|
| 458 |
+
"step": 30000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.26,
|
| 462 |
+
"learning_rate": 1.760876748540588e-05,
|
| 463 |
+
"loss": 0.2912,
|
| 464 |
+
"step": 31000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.26,
|
| 468 |
+
"eval_accuracy": 0.8708099847172694,
|
| 469 |
+
"eval_loss": 0.397215336561203,
|
| 470 |
+
"eval_runtime": 42.5529,
|
| 471 |
+
"eval_samples_per_second": 230.654,
|
| 472 |
+
"eval_steps_per_second": 28.835,
|
| 473 |
+
"step": 31000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.3,
|
| 477 |
+
"learning_rate": 1.7195726401586076e-05,
|
| 478 |
+
"loss": 0.2958,
|
| 479 |
+
"step": 32000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.3,
|
| 483 |
+
"eval_accuracy": 0.8731533367294957,
|
| 484 |
+
"eval_loss": 0.3713410496711731,
|
| 485 |
+
"eval_runtime": 42.8662,
|
| 486 |
+
"eval_samples_per_second": 228.968,
|
| 487 |
+
"eval_steps_per_second": 28.624,
|
| 488 |
+
"step": 32000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.34,
|
| 492 |
+
"learning_rate": 1.6782685317766274e-05,
|
| 493 |
+
"loss": 0.293,
|
| 494 |
+
"step": 33000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.34,
|
| 498 |
+
"eval_accuracy": 0.871523178807947,
|
| 499 |
+
"eval_loss": 0.3716830015182495,
|
| 500 |
+
"eval_runtime": 42.5402,
|
| 501 |
+
"eval_samples_per_second": 230.723,
|
| 502 |
+
"eval_steps_per_second": 28.843,
|
| 503 |
+
"step": 33000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.39,
|
| 507 |
+
"learning_rate": 1.636964423394647e-05,
|
| 508 |
+
"loss": 0.3001,
|
| 509 |
+
"step": 34000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.39,
|
| 513 |
+
"eval_accuracy": 0.8716250636780438,
|
| 514 |
+
"eval_loss": 0.3826219141483307,
|
| 515 |
+
"eval_runtime": 42.576,
|
| 516 |
+
"eval_samples_per_second": 230.529,
|
| 517 |
+
"eval_steps_per_second": 28.819,
|
| 518 |
+
"step": 34000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.43,
|
| 522 |
+
"learning_rate": 1.5956603150126667e-05,
|
| 523 |
+
"loss": 0.2864,
|
| 524 |
+
"step": 35000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.43,
|
| 528 |
+
"eval_accuracy": 0.8693835965359145,
|
| 529 |
+
"eval_loss": 0.41551369428634644,
|
| 530 |
+
"eval_runtime": 42.2749,
|
| 531 |
+
"eval_samples_per_second": 232.171,
|
| 532 |
+
"eval_steps_per_second": 29.024,
|
| 533 |
+
"step": 35000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.47,
|
| 537 |
+
"learning_rate": 1.554356206630686e-05,
|
| 538 |
+
"loss": 0.2827,
|
| 539 |
+
"step": 36000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.47,
|
| 543 |
+
"eval_accuracy": 0.866632705043301,
|
| 544 |
+
"eval_loss": 0.4223748743534088,
|
| 545 |
+
"eval_runtime": 42.4098,
|
| 546 |
+
"eval_samples_per_second": 231.432,
|
| 547 |
+
"eval_steps_per_second": 28.932,
|
| 548 |
+
"step": 36000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.51,
|
| 552 |
+
"learning_rate": 1.5130520982487058e-05,
|
| 553 |
+
"loss": 0.2836,
|
| 554 |
+
"step": 37000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.51,
|
| 558 |
+
"eval_accuracy": 0.8743759551706571,
|
| 559 |
+
"eval_loss": 0.3832014799118042,
|
| 560 |
+
"eval_runtime": 42.6737,
|
| 561 |
+
"eval_samples_per_second": 230.001,
|
| 562 |
+
"eval_steps_per_second": 28.753,
|
| 563 |
+
"step": 37000
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.55,
|
| 567 |
+
"learning_rate": 1.4717479898667254e-05,
|
| 568 |
+
"loss": 0.2844,
|
| 569 |
+
"step": 38000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.55,
|
| 573 |
+
"eval_accuracy": 0.8698930208863984,
|
| 574 |
+
"eval_loss": 0.4178868234157562,
|
| 575 |
+
"eval_runtime": 42.5391,
|
| 576 |
+
"eval_samples_per_second": 230.729,
|
| 577 |
+
"eval_steps_per_second": 28.844,
|
| 578 |
+
"step": 38000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.59,
|
| 582 |
+
"learning_rate": 1.430443881484745e-05,
|
| 583 |
+
"loss": 0.2866,
|
| 584 |
+
"step": 39000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.59,
|
| 588 |
+
"eval_accuracy": 0.8680590932246561,
|
| 589 |
+
"eval_loss": 0.3968764543533325,
|
| 590 |
+
"eval_runtime": 42.6173,
|
| 591 |
+
"eval_samples_per_second": 230.306,
|
| 592 |
+
"eval_steps_per_second": 28.791,
|
| 593 |
+
"step": 39000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.63,
|
| 597 |
+
"learning_rate": 1.3891397731027647e-05,
|
| 598 |
+
"loss": 0.2883,
|
| 599 |
+
"step": 40000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.63,
|
| 603 |
+
"eval_accuracy": 0.8682628629648497,
|
| 604 |
+
"eval_loss": 0.4000142514705658,
|
| 605 |
+
"eval_runtime": 42.4575,
|
| 606 |
+
"eval_samples_per_second": 231.172,
|
| 607 |
+
"eval_steps_per_second": 28.899,
|
| 608 |
+
"step": 40000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.67,
|
| 612 |
+
"learning_rate": 1.3478356647207844e-05,
|
| 613 |
+
"loss": 0.2832,
|
| 614 |
+
"step": 41000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.67,
|
| 618 |
+
"eval_accuracy": 0.8687722873153336,
|
| 619 |
+
"eval_loss": 0.3853473365306854,
|
| 620 |
+
"eval_runtime": 42.6575,
|
| 621 |
+
"eval_samples_per_second": 230.088,
|
| 622 |
+
"eval_steps_per_second": 28.764,
|
| 623 |
+
"step": 41000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.71,
|
| 627 |
+
"learning_rate": 1.3065315563388038e-05,
|
| 628 |
+
"loss": 0.2876,
|
| 629 |
+
"step": 42000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.71,
|
| 633 |
+
"eval_accuracy": 0.8676515537442689,
|
| 634 |
+
"eval_loss": 0.39242836833000183,
|
| 635 |
+
"eval_runtime": 42.1832,
|
| 636 |
+
"eval_samples_per_second": 232.675,
|
| 637 |
+
"eval_steps_per_second": 29.087,
|
| 638 |
+
"step": 42000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.75,
|
| 642 |
+
"learning_rate": 1.2652274479568235e-05,
|
| 643 |
+
"loss": 0.2855,
|
| 644 |
+
"step": 43000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.75,
|
| 648 |
+
"eval_accuracy": 0.8719307182883341,
|
| 649 |
+
"eval_loss": 0.4176536798477173,
|
| 650 |
+
"eval_runtime": 42.3591,
|
| 651 |
+
"eval_samples_per_second": 231.709,
|
| 652 |
+
"eval_steps_per_second": 28.967,
|
| 653 |
+
"step": 43000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.79,
|
| 657 |
+
"learning_rate": 1.2239233395748431e-05,
|
| 658 |
+
"loss": 0.2845,
|
| 659 |
+
"step": 44000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.79,
|
| 663 |
+
"eval_accuracy": 0.8724401426388181,
|
| 664 |
+
"eval_loss": 0.38765597343444824,
|
| 665 |
+
"eval_runtime": 42.3828,
|
| 666 |
+
"eval_samples_per_second": 231.58,
|
| 667 |
+
"eval_steps_per_second": 28.95,
|
| 668 |
+
"step": 44000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.83,
|
| 672 |
+
"learning_rate": 1.1826192311928628e-05,
|
| 673 |
+
"loss": 0.2882,
|
| 674 |
+
"step": 45000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.83,
|
| 678 |
+
"eval_accuracy": 0.8713194090677534,
|
| 679 |
+
"eval_loss": 0.3961141109466553,
|
| 680 |
+
"eval_runtime": 42.7036,
|
| 681 |
+
"eval_samples_per_second": 229.84,
|
| 682 |
+
"eval_steps_per_second": 28.733,
|
| 683 |
+
"step": 45000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.87,
|
| 687 |
+
"learning_rate": 1.1413151228108823e-05,
|
| 688 |
+
"loss": 0.2773,
|
| 689 |
+
"step": 46000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.87,
|
| 693 |
+
"eval_accuracy": 0.87396841569027,
|
| 694 |
+
"eval_loss": 0.37906670570373535,
|
| 695 |
+
"eval_runtime": 42.5306,
|
| 696 |
+
"eval_samples_per_second": 230.775,
|
| 697 |
+
"eval_steps_per_second": 28.85,
|
| 698 |
+
"step": 46000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.91,
|
| 702 |
+
"learning_rate": 1.1000110144289019e-05,
|
| 703 |
+
"loss": 0.2767,
|
| 704 |
+
"step": 47000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.91,
|
| 708 |
+
"eval_accuracy": 0.8779419256240448,
|
| 709 |
+
"eval_loss": 0.3877304494380951,
|
| 710 |
+
"eval_runtime": 42.632,
|
| 711 |
+
"eval_samples_per_second": 230.226,
|
| 712 |
+
"eval_steps_per_second": 28.781,
|
| 713 |
+
"step": 47000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.96,
|
| 717 |
+
"learning_rate": 1.0587069060469215e-05,
|
| 718 |
+
"loss": 0.2772,
|
| 719 |
+
"step": 48000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.96,
|
| 723 |
+
"eval_accuracy": 0.8689760570555273,
|
| 724 |
+
"eval_loss": 0.402159720659256,
|
| 725 |
+
"eval_runtime": 42.3751,
|
| 726 |
+
"eval_samples_per_second": 231.622,
|
| 727 |
+
"eval_steps_per_second": 28.956,
|
| 728 |
+
"step": 48000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.0,
|
| 732 |
+
"learning_rate": 1.017402797664941e-05,
|
| 733 |
+
"loss": 0.2816,
|
| 734 |
+
"step": 49000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.0,
|
| 738 |
+
"eval_accuracy": 0.8731533367294957,
|
| 739 |
+
"eval_loss": 0.3836565613746643,
|
| 740 |
+
"eval_runtime": 42.6247,
|
| 741 |
+
"eval_samples_per_second": 230.265,
|
| 742 |
+
"eval_steps_per_second": 28.786,
|
| 743 |
+
"step": 49000
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 2.04,
|
| 747 |
+
"learning_rate": 9.760986892829607e-06,
|
| 748 |
+
"loss": 0.2068,
|
| 749 |
+
"step": 50000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.04,
|
| 753 |
+
"eval_accuracy": 0.872032603158431,
|
| 754 |
+
"eval_loss": 0.4643520712852478,
|
| 755 |
+
"eval_runtime": 42.4953,
|
| 756 |
+
"eval_samples_per_second": 230.967,
|
| 757 |
+
"eval_steps_per_second": 28.874,
|
| 758 |
+
"step": 50000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.08,
|
| 762 |
+
"learning_rate": 9.347945809009803e-06,
|
| 763 |
+
"loss": 0.1914,
|
| 764 |
+
"step": 51000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.08,
|
| 768 |
+
"eval_accuracy": 0.8743759551706571,
|
| 769 |
+
"eval_loss": 0.49194175004959106,
|
| 770 |
+
"eval_runtime": 42.669,
|
| 771 |
+
"eval_samples_per_second": 230.027,
|
| 772 |
+
"eval_steps_per_second": 28.756,
|
| 773 |
+
"step": 51000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.12,
|
| 777 |
+
"learning_rate": 8.93490472519e-06,
|
| 778 |
+
"loss": 0.2,
|
| 779 |
+
"step": 52000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.12,
|
| 783 |
+
"eval_accuracy": 0.8701986754966887,
|
| 784 |
+
"eval_loss": 0.4869604706764221,
|
| 785 |
+
"eval_runtime": 42.5176,
|
| 786 |
+
"eval_samples_per_second": 230.845,
|
| 787 |
+
"eval_steps_per_second": 28.859,
|
| 788 |
+
"step": 52000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.16,
|
| 792 |
+
"learning_rate": 8.521863641370194e-06,
|
| 793 |
+
"loss": 0.1904,
|
| 794 |
+
"step": 53000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.16,
|
| 798 |
+
"eval_accuracy": 0.8736627610799796,
|
| 799 |
+
"eval_loss": 0.5038197636604309,
|
| 800 |
+
"eval_runtime": 42.6855,
|
| 801 |
+
"eval_samples_per_second": 229.937,
|
| 802 |
+
"eval_steps_per_second": 28.745,
|
| 803 |
+
"step": 53000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.2,
|
| 807 |
+
"learning_rate": 8.10882255755039e-06,
|
| 808 |
+
"loss": 0.1915,
|
| 809 |
+
"step": 54000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.2,
|
| 813 |
+
"eval_accuracy": 0.8711156393275599,
|
| 814 |
+
"eval_loss": 0.5232452750205994,
|
| 815 |
+
"eval_runtime": 42.4057,
|
| 816 |
+
"eval_samples_per_second": 231.455,
|
| 817 |
+
"eval_steps_per_second": 28.935,
|
| 818 |
+
"step": 54000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.24,
|
| 822 |
+
"learning_rate": 7.695781473730587e-06,
|
| 823 |
+
"loss": 0.1956,
|
| 824 |
+
"step": 55000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.24,
|
| 828 |
+
"eval_accuracy": 0.8746816097809476,
|
| 829 |
+
"eval_loss": 0.5192070603370667,
|
| 830 |
+
"eval_runtime": 42.5451,
|
| 831 |
+
"eval_samples_per_second": 230.697,
|
| 832 |
+
"eval_steps_per_second": 28.84,
|
| 833 |
+
"step": 55000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.28,
|
| 837 |
+
"learning_rate": 7.282740389910783e-06,
|
| 838 |
+
"loss": 0.1911,
|
| 839 |
+
"step": 56000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.28,
|
| 843 |
+
"eval_accuracy": 0.8761079979623025,
|
| 844 |
+
"eval_loss": 0.5215316414833069,
|
| 845 |
+
"eval_runtime": 42.8008,
|
| 846 |
+
"eval_samples_per_second": 229.318,
|
| 847 |
+
"eval_steps_per_second": 28.668,
|
| 848 |
+
"step": 56000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.32,
|
| 852 |
+
"learning_rate": 6.869699306090979e-06,
|
| 853 |
+
"loss": 0.2053,
|
| 854 |
+
"step": 57000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.32,
|
| 858 |
+
"eval_accuracy": 0.8737646459500764,
|
| 859 |
+
"eval_loss": 0.460406094789505,
|
| 860 |
+
"eval_runtime": 42.6406,
|
| 861 |
+
"eval_samples_per_second": 230.18,
|
| 862 |
+
"eval_steps_per_second": 28.775,
|
| 863 |
+
"step": 57000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.36,
|
| 867 |
+
"learning_rate": 6.456658222271175e-06,
|
| 868 |
+
"loss": 0.2008,
|
| 869 |
+
"step": 58000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.36,
|
| 873 |
+
"eval_accuracy": 0.871523178807947,
|
| 874 |
+
"eval_loss": 0.5162080526351929,
|
| 875 |
+
"eval_runtime": 42.396,
|
| 876 |
+
"eval_samples_per_second": 231.508,
|
| 877 |
+
"eval_steps_per_second": 28.941,
|
| 878 |
+
"step": 58000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.4,
|
| 882 |
+
"learning_rate": 6.043617138451371e-06,
|
| 883 |
+
"loss": 0.1971,
|
| 884 |
+
"step": 59000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.4,
|
| 888 |
+
"eval_accuracy": 0.875394803871625,
|
| 889 |
+
"eval_loss": 0.4885903000831604,
|
| 890 |
+
"eval_runtime": 42.2887,
|
| 891 |
+
"eval_samples_per_second": 232.095,
|
| 892 |
+
"eval_steps_per_second": 29.015,
|
| 893 |
+
"step": 59000
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.44,
|
| 897 |
+
"learning_rate": 5.630576054631567e-06,
|
| 898 |
+
"loss": 0.192,
|
| 899 |
+
"step": 60000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.44,
|
| 903 |
+
"eval_accuracy": 0.872542027508915,
|
| 904 |
+
"eval_loss": 0.49207794666290283,
|
| 905 |
+
"eval_runtime": 42.6387,
|
| 906 |
+
"eval_samples_per_second": 230.19,
|
| 907 |
+
"eval_steps_per_second": 28.777,
|
| 908 |
+
"step": 60000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.49,
|
| 912 |
+
"learning_rate": 5.217534970811764e-06,
|
| 913 |
+
"loss": 0.1937,
|
| 914 |
+
"step": 61000
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.49,
|
| 918 |
+
"eval_accuracy": 0.8763117677024962,
|
| 919 |
+
"eval_loss": 0.4916786253452301,
|
| 920 |
+
"eval_runtime": 42.3165,
|
| 921 |
+
"eval_samples_per_second": 231.943,
|
| 922 |
+
"eval_steps_per_second": 28.996,
|
| 923 |
+
"step": 61000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.53,
|
| 927 |
+
"learning_rate": 4.80449388699196e-06,
|
| 928 |
+
"loss": 0.1931,
|
| 929 |
+
"step": 62000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 2.53,
|
| 933 |
+
"eval_accuracy": 0.8778400407539481,
|
| 934 |
+
"eval_loss": 0.47893819212913513,
|
| 935 |
+
"eval_runtime": 42.4257,
|
| 936 |
+
"eval_samples_per_second": 231.346,
|
| 937 |
+
"eval_steps_per_second": 28.921,
|
| 938 |
+
"step": 62000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.57,
|
| 942 |
+
"learning_rate": 4.391452803172156e-06,
|
| 943 |
+
"loss": 0.1964,
|
| 944 |
+
"step": 63000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.57,
|
| 948 |
+
"eval_accuracy": 0.8721344880285278,
|
| 949 |
+
"eval_loss": 0.49971821904182434,
|
| 950 |
+
"eval_runtime": 42.5737,
|
| 951 |
+
"eval_samples_per_second": 230.542,
|
| 952 |
+
"eval_steps_per_second": 28.821,
|
| 953 |
+
"step": 63000
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 2.61,
|
| 957 |
+
"learning_rate": 3.978411719352352e-06,
|
| 958 |
+
"loss": 0.2008,
|
| 959 |
+
"step": 64000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.61,
|
| 963 |
+
"eval_accuracy": 0.8755985736118187,
|
| 964 |
+
"eval_loss": 0.4747646749019623,
|
| 965 |
+
"eval_runtime": 42.3736,
|
| 966 |
+
"eval_samples_per_second": 231.63,
|
| 967 |
+
"eval_steps_per_second": 28.957,
|
| 968 |
+
"step": 64000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.65,
|
| 972 |
+
"learning_rate": 3.5653706355325475e-06,
|
| 973 |
+
"loss": 0.1962,
|
| 974 |
+
"step": 65000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 2.65,
|
| 978 |
+
"eval_accuracy": 0.876413652572593,
|
| 979 |
+
"eval_loss": 0.4839693605899811,
|
| 980 |
+
"eval_runtime": 42.5306,
|
| 981 |
+
"eval_samples_per_second": 230.775,
|
| 982 |
+
"eval_steps_per_second": 28.85,
|
| 983 |
+
"step": 65000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.69,
|
| 987 |
+
"learning_rate": 3.1523295517127435e-06,
|
| 988 |
+
"loss": 0.2029,
|
| 989 |
+
"step": 66000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.69,
|
| 993 |
+
"eval_accuracy": 0.8767193071828834,
|
| 994 |
+
"eval_loss": 0.48892590403556824,
|
| 995 |
+
"eval_runtime": 42.2422,
|
| 996 |
+
"eval_samples_per_second": 232.351,
|
| 997 |
+
"eval_steps_per_second": 29.047,
|
| 998 |
+
"step": 66000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.73,
|
| 1002 |
+
"learning_rate": 2.73928846789294e-06,
|
| 1003 |
+
"loss": 0.1927,
|
| 1004 |
+
"step": 67000
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.73,
|
| 1008 |
+
"eval_accuracy": 0.8758023433520122,
|
| 1009 |
+
"eval_loss": 0.4820311963558197,
|
| 1010 |
+
"eval_runtime": 42.6573,
|
| 1011 |
+
"eval_samples_per_second": 230.089,
|
| 1012 |
+
"eval_steps_per_second": 28.764,
|
| 1013 |
+
"step": 67000
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.77,
|
| 1017 |
+
"learning_rate": 2.326247384073136e-06,
|
| 1018 |
+
"loss": 0.1926,
|
| 1019 |
+
"step": 68000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.77,
|
| 1023 |
+
"eval_accuracy": 0.8762098828323994,
|
| 1024 |
+
"eval_loss": 0.48573482036590576,
|
| 1025 |
+
"eval_runtime": 42.7141,
|
| 1026 |
+
"eval_samples_per_second": 229.783,
|
| 1027 |
+
"eval_steps_per_second": 28.726,
|
| 1028 |
+
"step": 68000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.81,
|
| 1032 |
+
"learning_rate": 1.913206300253332e-06,
|
| 1033 |
+
"loss": 0.1919,
|
| 1034 |
+
"step": 69000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 2.81,
|
| 1038 |
+
"eval_accuracy": 0.8748853795211411,
|
| 1039 |
+
"eval_loss": 0.4835805296897888,
|
| 1040 |
+
"eval_runtime": 42.3433,
|
| 1041 |
+
"eval_samples_per_second": 231.796,
|
| 1042 |
+
"eval_steps_per_second": 28.977,
|
| 1043 |
+
"step": 69000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.85,
|
| 1047 |
+
"learning_rate": 1.500165216433528e-06,
|
| 1048 |
+
"loss": 0.1911,
|
| 1049 |
+
"step": 70000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.85,
|
| 1053 |
+
"eval_accuracy": 0.8741721854304636,
|
| 1054 |
+
"eval_loss": 0.48588985204696655,
|
| 1055 |
+
"eval_runtime": 42.2277,
|
| 1056 |
+
"eval_samples_per_second": 232.43,
|
| 1057 |
+
"eval_steps_per_second": 29.057,
|
| 1058 |
+
"step": 70000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.89,
|
| 1062 |
+
"learning_rate": 1.087124132613724e-06,
|
| 1063 |
+
"loss": 0.1897,
|
| 1064 |
+
"step": 71000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.89,
|
| 1068 |
+
"eval_accuracy": 0.8766174223127865,
|
| 1069 |
+
"eval_loss": 0.48527196049690247,
|
| 1070 |
+
"eval_runtime": 42.5077,
|
| 1071 |
+
"eval_samples_per_second": 230.899,
|
| 1072 |
+
"eval_steps_per_second": 28.865,
|
| 1073 |
+
"step": 71000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.93,
|
| 1077 |
+
"learning_rate": 6.740830487939201e-07,
|
| 1078 |
+
"loss": 0.186,
|
| 1079 |
+
"step": 72000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.93,
|
| 1083 |
+
"eval_accuracy": 0.8768211920529801,
|
| 1084 |
+
"eval_loss": 0.4946127235889435,
|
| 1085 |
+
"eval_runtime": 42.345,
|
| 1086 |
+
"eval_samples_per_second": 231.786,
|
| 1087 |
+
"eval_steps_per_second": 28.976,
|
| 1088 |
+
"step": 72000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.97,
|
| 1092 |
+
"learning_rate": 2.610419649741161e-07,
|
| 1093 |
+
"loss": 0.2011,
|
| 1094 |
+
"step": 73000
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.97,
|
| 1098 |
+
"eval_accuracy": 0.8767193071828834,
|
| 1099 |
+
"eval_loss": 0.4851147532463074,
|
| 1100 |
+
"eval_runtime": 42.2673,
|
| 1101 |
+
"eval_samples_per_second": 232.213,
|
| 1102 |
+
"eval_steps_per_second": 29.03,
|
| 1103 |
+
"step": 73000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 3.0,
|
| 1107 |
+
"step": 73632,
|
| 1108 |
+
"total_flos": 7.803448964124365e+16,
|
| 1109 |
+
"train_loss": 0.3087343405972663,
|
| 1110 |
+
"train_runtime": 16507.3954,
|
| 1111 |
+
"train_samples_per_second": 71.368,
|
| 1112 |
+
"train_steps_per_second": 4.461
|
| 1113 |
+
}
|
| 1114 |
+
],
|
| 1115 |
+
"max_steps": 73632,
|
| 1116 |
+
"num_train_epochs": 3,
|
| 1117 |
+
"total_flos": 7.803448964124365e+16,
|
| 1118 |
+
"trial_name": null,
|
| 1119 |
+
"trial_params": null
|
| 1120 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a16a9c294c00e988e8145e80b5a15ec23ce646263c19bb77c2d615439e6186a9
|
| 3 |
+
size 2799
|