Upload 17 files
Browse files- README.md +158 -1
- logs/bg_cn_lang_adapter.png +0 -0
- logs/events.out.tfevents.1709072592.serv-3318.1735865.0 +3 -0
- logs/events.out.tfevents.1709075335.serv-3318.1735865.1 +3 -0
- mlm/adapter_config.json +41 -0
- mlm/head_config.json +14 -0
- mlm/pytorch_adapter.bin +3 -0
- mlm/pytorch_model_head.bin +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +1099 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
CHANGED
|
@@ -1,3 +1,160 @@
|
|
| 1 |
---
|
| 2 |
-
license:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: bert-base-multilingual-cased
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: bg
|
| 10 |
+
results: []
|
| 11 |
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# bg
|
| 17 |
+
|
| 18 |
+
This model is an adapter fine-tuned on top of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on the Bulgarian ConceptNet dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.4640
|
| 21 |
+
- Accuracy: 0.8875
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 16
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- training_steps: 50000
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 53 |
+
| 1.5057 | 0.15 | 500 | 0.9846 | 0.8149 |
|
| 54 |
+
| 1.0172 | 0.31 | 1000 | 0.8395 | 0.8259 |
|
| 55 |
+
| 0.8814 | 0.46 | 1500 | 0.7823 | 0.8368 |
|
| 56 |
+
| 0.8405 | 0.61 | 2000 | 0.7437 | 0.8449 |
|
| 57 |
+
| 0.7773 | 0.77 | 2500 | 0.7247 | 0.8387 |
|
| 58 |
+
| 0.7762 | 0.92 | 3000 | 0.6521 | 0.8513 |
|
| 59 |
+
| 0.7186 | 1.07 | 3500 | 0.6834 | 0.8492 |
|
| 60 |
+
| 0.7033 | 1.22 | 4000 | 0.6715 | 0.8523 |
|
| 61 |
+
| 0.672 | 1.38 | 4500 | 0.6539 | 0.8560 |
|
| 62 |
+
| 0.6613 | 1.53 | 5000 | 0.6387 | 0.8567 |
|
| 63 |
+
| 0.6712 | 1.68 | 5500 | 0.6180 | 0.8624 |
|
| 64 |
+
| 0.6776 | 1.84 | 6000 | 0.6635 | 0.8537 |
|
| 65 |
+
| 0.6484 | 1.99 | 6500 | 0.5946 | 0.8661 |
|
| 66 |
+
| 0.6817 | 2.14 | 7000 | 0.6126 | 0.8655 |
|
| 67 |
+
| 0.6392 | 2.3 | 7500 | 0.6136 | 0.8613 |
|
| 68 |
+
| 0.6394 | 2.45 | 8000 | 0.6321 | 0.8621 |
|
| 69 |
+
| 0.6273 | 2.6 | 8500 | 0.5997 | 0.8629 |
|
| 70 |
+
| 0.5993 | 2.76 | 9000 | 0.6028 | 0.8646 |
|
| 71 |
+
| 0.6527 | 2.91 | 9500 | 0.6584 | 0.8510 |
|
| 72 |
+
| 0.5897 | 3.06 | 10000 | 0.5728 | 0.8676 |
|
| 73 |
+
| 0.574 | 3.21 | 10500 | 0.5870 | 0.8671 |
|
| 74 |
+
| 0.6026 | 3.37 | 11000 | 0.6067 | 0.8677 |
|
| 75 |
+
| 0.5896 | 3.52 | 11500 | 0.6000 | 0.8638 |
|
| 76 |
+
| 0.566 | 3.67 | 12000 | 0.5566 | 0.8712 |
|
| 77 |
+
| 0.5928 | 3.83 | 12500 | 0.5621 | 0.8675 |
|
| 78 |
+
| 0.597 | 3.98 | 13000 | 0.5162 | 0.8771 |
|
| 79 |
+
| 0.5836 | 4.13 | 13500 | 0.5498 | 0.8696 |
|
| 80 |
+
| 0.5864 | 4.29 | 14000 | 0.5728 | 0.8640 |
|
| 81 |
+
| 0.5562 | 4.44 | 14500 | 0.6000 | 0.8623 |
|
| 82 |
+
| 0.5999 | 4.59 | 15000 | 0.5589 | 0.8679 |
|
| 83 |
+
| 0.5767 | 4.75 | 15500 | 0.5713 | 0.8681 |
|
| 84 |
+
| 0.5574 | 4.9 | 16000 | 0.5338 | 0.8739 |
|
| 85 |
+
| 0.568 | 5.05 | 16500 | 0.5527 | 0.8725 |
|
| 86 |
+
| 0.5568 | 5.21 | 17000 | 0.5058 | 0.8777 |
|
| 87 |
+
| 0.5369 | 5.36 | 17500 | 0.5599 | 0.8720 |
|
| 88 |
+
| 0.518 | 5.51 | 18000 | 0.5610 | 0.8720 |
|
| 89 |
+
| 0.5637 | 5.66 | 18500 | 0.5467 | 0.8728 |
|
| 90 |
+
| 0.557 | 5.82 | 19000 | 0.5349 | 0.8714 |
|
| 91 |
+
| 0.5499 | 5.97 | 19500 | 0.5468 | 0.8724 |
|
| 92 |
+
| 0.5304 | 6.12 | 20000 | 0.5243 | 0.8741 |
|
| 93 |
+
| 0.5431 | 6.28 | 20500 | 0.4998 | 0.8784 |
|
| 94 |
+
| 0.5508 | 6.43 | 21000 | 0.5367 | 0.8764 |
|
| 95 |
+
| 0.5701 | 6.58 | 21500 | 0.5365 | 0.8734 |
|
| 96 |
+
| 0.521 | 6.74 | 22000 | 0.4879 | 0.8819 |
|
| 97 |
+
| 0.5514 | 6.89 | 22500 | 0.5106 | 0.8787 |
|
| 98 |
+
| 0.547 | 7.04 | 23000 | 0.5258 | 0.8747 |
|
| 99 |
+
| 0.5512 | 7.2 | 23500 | 0.4975 | 0.8778 |
|
| 100 |
+
| 0.5407 | 7.35 | 24000 | 0.4944 | 0.8786 |
|
| 101 |
+
| 0.5181 | 7.5 | 24500 | 0.4912 | 0.8795 |
|
| 102 |
+
| 0.5493 | 7.65 | 25000 | 0.5188 | 0.8730 |
|
| 103 |
+
| 0.5388 | 7.81 | 25500 | 0.5000 | 0.8831 |
|
| 104 |
+
| 0.5284 | 7.96 | 26000 | 0.5161 | 0.8737 |
|
| 105 |
+
| 0.5116 | 8.11 | 26500 | 0.5263 | 0.8760 |
|
| 106 |
+
| 0.5161 | 8.27 | 27000 | 0.5002 | 0.8787 |
|
| 107 |
+
| 0.5185 | 8.42 | 27500 | 0.5127 | 0.8745 |
|
| 108 |
+
| 0.5291 | 8.57 | 28000 | 0.5116 | 0.8782 |
|
| 109 |
+
| 0.5061 | 8.73 | 28500 | 0.4972 | 0.8774 |
|
| 110 |
+
| 0.479 | 8.88 | 29000 | 0.4978 | 0.8798 |
|
| 111 |
+
| 0.5154 | 9.03 | 29500 | 0.5088 | 0.8771 |
|
| 112 |
+
| 0.4989 | 9.19 | 30000 | 0.5119 | 0.8744 |
|
| 113 |
+
| 0.5098 | 9.34 | 30500 | 0.4916 | 0.8826 |
|
| 114 |
+
| 0.4777 | 9.49 | 31000 | 0.4957 | 0.8824 |
|
| 115 |
+
| 0.5462 | 9.64 | 31500 | 0.4846 | 0.8779 |
|
| 116 |
+
| 0.509 | 9.8 | 32000 | 0.4873 | 0.8810 |
|
| 117 |
+
| 0.5181 | 9.95 | 32500 | 0.5227 | 0.8710 |
|
| 118 |
+
| 0.5269 | 10.1 | 33000 | 0.4929 | 0.8803 |
|
| 119 |
+
| 0.5094 | 10.26 | 33500 | 0.4841 | 0.8877 |
|
| 120 |
+
| 0.5033 | 10.41 | 34000 | 0.5129 | 0.8805 |
|
| 121 |
+
| 0.4913 | 10.56 | 34500 | 0.4978 | 0.8789 |
|
| 122 |
+
| 0.4938 | 10.72 | 35000 | 0.4640 | 0.8838 |
|
| 123 |
+
| 0.4954 | 10.87 | 35500 | 0.4991 | 0.8794 |
|
| 124 |
+
| 0.458 | 11.02 | 36000 | 0.4453 | 0.8886 |
|
| 125 |
+
| 0.526 | 11.18 | 36500 | 0.4863 | 0.8832 |
|
| 126 |
+
| 0.4809 | 11.33 | 37000 | 0.4923 | 0.8784 |
|
| 127 |
+
| 0.466 | 11.48 | 37500 | 0.4824 | 0.8807 |
|
| 128 |
+
| 0.4903 | 11.64 | 38000 | 0.4552 | 0.8848 |
|
| 129 |
+
| 0.4875 | 11.79 | 38500 | 0.4850 | 0.8780 |
|
| 130 |
+
| 0.4858 | 11.94 | 39000 | 0.4728 | 0.8833 |
|
| 131 |
+
| 0.4868 | 12.09 | 39500 | 0.4868 | 0.8800 |
|
| 132 |
+
| 0.485 | 12.25 | 40000 | 0.4935 | 0.8802 |
|
| 133 |
+
| 0.4823 | 12.4 | 40500 | 0.4789 | 0.8828 |
|
| 134 |
+
| 0.4629 | 12.55 | 41000 | 0.4834 | 0.8835 |
|
| 135 |
+
| 0.4915 | 12.71 | 41500 | 0.4864 | 0.8812 |
|
| 136 |
+
| 0.473 | 12.86 | 42000 | 0.5136 | 0.8793 |
|
| 137 |
+
| 0.4849 | 13.01 | 42500 | 0.4823 | 0.8815 |
|
| 138 |
+
| 0.4582 | 13.17 | 43000 | 0.4637 | 0.8844 |
|
| 139 |
+
| 0.4938 | 13.32 | 43500 | 0.4829 | 0.8842 |
|
| 140 |
+
| 0.4682 | 13.47 | 44000 | 0.4799 | 0.8817 |
|
| 141 |
+
| 0.4885 | 13.63 | 44500 | 0.4754 | 0.8858 |
|
| 142 |
+
| 0.4641 | 13.78 | 45000 | 0.4738 | 0.8849 |
|
| 143 |
+
| 0.4664 | 13.93 | 45500 | 0.4512 | 0.8869 |
|
| 144 |
+
| 0.4722 | 14.08 | 46000 | 0.4821 | 0.8836 |
|
| 145 |
+
| 0.485 | 14.24 | 46500 | 0.4735 | 0.8842 |
|
| 146 |
+
| 0.4784 | 14.39 | 47000 | 0.4557 | 0.8823 |
|
| 147 |
+
| 0.4821 | 14.54 | 47500 | 0.4707 | 0.8856 |
|
| 148 |
+
| 0.478 | 14.7 | 48000 | 0.4682 | 0.8846 |
|
| 149 |
+
| 0.451 | 14.85 | 48500 | 0.4744 | 0.8781 |
|
| 150 |
+
| 0.4582 | 15.0 | 49000 | 0.4617 | 0.8835 |
|
| 151 |
+
| 0.4949 | 15.16 | 49500 | 0.4769 | 0.8835 |
|
| 152 |
+
| 0.4546 | 15.31 | 50000 | 0.4677 | 0.8835 |
|
| 153 |
+
|
| 154 |
+
|
| 155 |
+
### Framework versions
|
| 156 |
+
|
| 157 |
+
- Transformers 4.35.2
|
| 158 |
+
- Pytorch 2.0.0
|
| 159 |
+
- Datasets 2.15.0
|
| 160 |
+
- Tokenizers 0.15.0
|
logs/bg_cn_lang_adapter.png
ADDED
|
logs/events.out.tfevents.1709072592.serv-3318.1735865.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bc37bc50f2d5e69e98e5753b90f687e311a30a43be901b779b4ba5d484b4e33
|
| 3 |
+
size 53301
|
logs/events.out.tfevents.1709075335.serv-3318.1735865.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bed101833d8c96a8c8f862a883435f5519523000c2d6bcac36446bcbe9872d0
|
| 3 |
+
size 369
|
mlm/adapter_config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"adapter_residual_before_ln": false,
|
| 4 |
+
"cross_adapter": false,
|
| 5 |
+
"factorized_phm_W": true,
|
| 6 |
+
"factorized_phm_rule": false,
|
| 7 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 8 |
+
"init_weights": "bert",
|
| 9 |
+
"inv_adapter": null,
|
| 10 |
+
"inv_adapter_reduction_factor": null,
|
| 11 |
+
"is_parallel": false,
|
| 12 |
+
"learn_phm": true,
|
| 13 |
+
"leave_out": [],
|
| 14 |
+
"ln_after": false,
|
| 15 |
+
"ln_before": false,
|
| 16 |
+
"mh_adapter": false,
|
| 17 |
+
"non_linearity": "relu",
|
| 18 |
+
"original_ln_after": true,
|
| 19 |
+
"original_ln_before": true,
|
| 20 |
+
"output_adapter": true,
|
| 21 |
+
"phm_bias": true,
|
| 22 |
+
"phm_c_init": "normal",
|
| 23 |
+
"phm_dim": 4,
|
| 24 |
+
"phm_init_range": 0.0001,
|
| 25 |
+
"phm_layer": false,
|
| 26 |
+
"phm_rank": 1,
|
| 27 |
+
"reduction_factor": 16,
|
| 28 |
+
"residual_before_ln": true,
|
| 29 |
+
"scaling": 1.0,
|
| 30 |
+
"shared_W_phm": false,
|
| 31 |
+
"shared_phm_rule": true,
|
| 32 |
+
"use_gating": false
|
| 33 |
+
},
|
| 34 |
+
"config_id": "9076f36a74755ac4",
|
| 35 |
+
"hidden_size": 768,
|
| 36 |
+
"model_class": "BertForMaskedLM",
|
| 37 |
+
"model_name": "bert-base-multilingual-cased",
|
| 38 |
+
"model_type": "bert",
|
| 39 |
+
"name": "mlm",
|
| 40 |
+
"version": "0.1.1"
|
| 41 |
+
}
|
mlm/head_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 768,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1
|
| 7 |
+
},
|
| 8 |
+
"model_class": "BertForMaskedLM",
|
| 9 |
+
"model_name": "bert-base-multilingual-cased",
|
| 10 |
+
"model_type": "bert",
|
| 11 |
+
"name": null,
|
| 12 |
+
"num_labels": 2,
|
| 13 |
+
"version": "0.1.1"
|
| 14 |
+
}
|
mlm/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b223a57f41a9e6a2c8debf7d0788b91e126ddb1622be27f1c76edade60b5a31b
|
| 3 |
+
size 3594917
|
mlm/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ad03a83a7dae20b6f0cbf36a8063b17d8a1bc9c311e1aaa69385d1b16a8066b
|
| 3 |
+
size 370097519
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5961573ed8d14c1f436754dd9784d611c4c59aca2b69b6533123ff92fee84ee
|
| 3 |
+
size 11936581
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:373c00f5caea17a3661aaf0bf53fb559e5d31184ff8fa45fd88e5a14d98c6dbb
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a74271501810f2b6d0d27ae7c5e032ef1ab90b92dac7c1a79a051174b3fa8163
|
| 3 |
+
size 627
|
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 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": false,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1099 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4452793300151825,
|
| 3 |
+
"best_model_checkpoint": "./models/adapters_mlm_cn/bg/checkpoint-36000",
|
| 4 |
+
"epoch": 11.022657685241887,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 36000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.15,
|
| 13 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 14 |
+
"loss": 1.5057,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.15,
|
| 19 |
+
"eval_accuracy": 0.8148620791204476,
|
| 20 |
+
"eval_loss": 0.9846400618553162,
|
| 21 |
+
"eval_runtime": 7.7173,
|
| 22 |
+
"eval_samples_per_second": 752.339,
|
| 23 |
+
"eval_steps_per_second": 47.037,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.31,
|
| 28 |
+
"learning_rate": 4.9e-05,
|
| 29 |
+
"loss": 1.0172,
|
| 30 |
+
"step": 1000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.31,
|
| 34 |
+
"eval_accuracy": 0.82587890625,
|
| 35 |
+
"eval_loss": 0.8394753932952881,
|
| 36 |
+
"eval_runtime": 7.7319,
|
| 37 |
+
"eval_samples_per_second": 750.918,
|
| 38 |
+
"eval_steps_per_second": 46.949,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.46,
|
| 43 |
+
"learning_rate": 4.85e-05,
|
| 44 |
+
"loss": 0.8814,
|
| 45 |
+
"step": 1500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.46,
|
| 49 |
+
"eval_accuracy": 0.8368038740920097,
|
| 50 |
+
"eval_loss": 0.7822620272636414,
|
| 51 |
+
"eval_runtime": 7.7294,
|
| 52 |
+
"eval_samples_per_second": 751.162,
|
| 53 |
+
"eval_steps_per_second": 46.964,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.61,
|
| 58 |
+
"learning_rate": 4.8e-05,
|
| 59 |
+
"loss": 0.8405,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.61,
|
| 64 |
+
"eval_accuracy": 0.8449071800412533,
|
| 65 |
+
"eval_loss": 0.7436666488647461,
|
| 66 |
+
"eval_runtime": 7.9259,
|
| 67 |
+
"eval_samples_per_second": 732.538,
|
| 68 |
+
"eval_steps_per_second": 45.799,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.77,
|
| 73 |
+
"learning_rate": 4.75e-05,
|
| 74 |
+
"loss": 0.7773,
|
| 75 |
+
"step": 2500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.77,
|
| 79 |
+
"eval_accuracy": 0.8386841062227507,
|
| 80 |
+
"eval_loss": 0.7246997952461243,
|
| 81 |
+
"eval_runtime": 7.7331,
|
| 82 |
+
"eval_samples_per_second": 750.803,
|
| 83 |
+
"eval_steps_per_second": 46.941,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.92,
|
| 88 |
+
"learning_rate": 4.7e-05,
|
| 89 |
+
"loss": 0.7762,
|
| 90 |
+
"step": 3000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.92,
|
| 94 |
+
"eval_accuracy": 0.8513044340839202,
|
| 95 |
+
"eval_loss": 0.6520901322364807,
|
| 96 |
+
"eval_runtime": 7.9369,
|
| 97 |
+
"eval_samples_per_second": 731.516,
|
| 98 |
+
"eval_steps_per_second": 45.736,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.07,
|
| 103 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 104 |
+
"loss": 0.7186,
|
| 105 |
+
"step": 3500
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.07,
|
| 109 |
+
"eval_accuracy": 0.8492265517916585,
|
| 110 |
+
"eval_loss": 0.6834315061569214,
|
| 111 |
+
"eval_runtime": 7.6738,
|
| 112 |
+
"eval_samples_per_second": 756.596,
|
| 113 |
+
"eval_steps_per_second": 47.304,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.22,
|
| 118 |
+
"learning_rate": 4.600000000000001e-05,
|
| 119 |
+
"loss": 0.7033,
|
| 120 |
+
"step": 4000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.22,
|
| 124 |
+
"eval_accuracy": 0.852271607371637,
|
| 125 |
+
"eval_loss": 0.67154860496521,
|
| 126 |
+
"eval_runtime": 7.7294,
|
| 127 |
+
"eval_samples_per_second": 751.163,
|
| 128 |
+
"eval_steps_per_second": 46.964,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.38,
|
| 133 |
+
"learning_rate": 4.55e-05,
|
| 134 |
+
"loss": 0.672,
|
| 135 |
+
"step": 4500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.38,
|
| 139 |
+
"eval_accuracy": 0.855973974763407,
|
| 140 |
+
"eval_loss": 0.6539207100868225,
|
| 141 |
+
"eval_runtime": 7.7117,
|
| 142 |
+
"eval_samples_per_second": 752.881,
|
| 143 |
+
"eval_steps_per_second": 47.071,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.53,
|
| 148 |
+
"learning_rate": 4.5e-05,
|
| 149 |
+
"loss": 0.6613,
|
| 150 |
+
"step": 5000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.53,
|
| 154 |
+
"eval_accuracy": 0.8567085131424088,
|
| 155 |
+
"eval_loss": 0.638721227645874,
|
| 156 |
+
"eval_runtime": 7.6505,
|
| 157 |
+
"eval_samples_per_second": 758.9,
|
| 158 |
+
"eval_steps_per_second": 47.448,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.68,
|
| 163 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 164 |
+
"loss": 0.6712,
|
| 165 |
+
"step": 5500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.68,
|
| 169 |
+
"eval_accuracy": 0.862372613040467,
|
| 170 |
+
"eval_loss": 0.6180465221405029,
|
| 171 |
+
"eval_runtime": 7.7012,
|
| 172 |
+
"eval_samples_per_second": 753.913,
|
| 173 |
+
"eval_steps_per_second": 47.136,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.84,
|
| 178 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 179 |
+
"loss": 0.6776,
|
| 180 |
+
"step": 6000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.84,
|
| 184 |
+
"eval_accuracy": 0.8537038849202466,
|
| 185 |
+
"eval_loss": 0.6634594202041626,
|
| 186 |
+
"eval_runtime": 7.7042,
|
| 187 |
+
"eval_samples_per_second": 753.61,
|
| 188 |
+
"eval_steps_per_second": 47.117,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.99,
|
| 193 |
+
"learning_rate": 4.35e-05,
|
| 194 |
+
"loss": 0.6484,
|
| 195 |
+
"step": 6500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.99,
|
| 199 |
+
"eval_accuracy": 0.8661394258933802,
|
| 200 |
+
"eval_loss": 0.5945894122123718,
|
| 201 |
+
"eval_runtime": 7.6974,
|
| 202 |
+
"eval_samples_per_second": 754.283,
|
| 203 |
+
"eval_steps_per_second": 47.159,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.14,
|
| 208 |
+
"learning_rate": 4.3e-05,
|
| 209 |
+
"loss": 0.6817,
|
| 210 |
+
"step": 7000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.14,
|
| 214 |
+
"eval_accuracy": 0.8654563297350344,
|
| 215 |
+
"eval_loss": 0.6126104593276978,
|
| 216 |
+
"eval_runtime": 8.509,
|
| 217 |
+
"eval_samples_per_second": 682.334,
|
| 218 |
+
"eval_steps_per_second": 42.661,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.3,
|
| 223 |
+
"learning_rate": 4.25e-05,
|
| 224 |
+
"loss": 0.6392,
|
| 225 |
+
"step": 7500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.3,
|
| 229 |
+
"eval_accuracy": 0.8613216715257531,
|
| 230 |
+
"eval_loss": 0.613590657711029,
|
| 231 |
+
"eval_runtime": 8.1378,
|
| 232 |
+
"eval_samples_per_second": 713.458,
|
| 233 |
+
"eval_steps_per_second": 44.606,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.45,
|
| 238 |
+
"learning_rate": 4.2e-05,
|
| 239 |
+
"loss": 0.6394,
|
| 240 |
+
"step": 8000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.45,
|
| 244 |
+
"eval_accuracy": 0.8620723749258453,
|
| 245 |
+
"eval_loss": 0.6320650577545166,
|
| 246 |
+
"eval_runtime": 7.7697,
|
| 247 |
+
"eval_samples_per_second": 747.26,
|
| 248 |
+
"eval_steps_per_second": 46.72,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.6,
|
| 253 |
+
"learning_rate": 4.15e-05,
|
| 254 |
+
"loss": 0.6273,
|
| 255 |
+
"step": 8500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.6,
|
| 259 |
+
"eval_accuracy": 0.8629402009560043,
|
| 260 |
+
"eval_loss": 0.5997043251991272,
|
| 261 |
+
"eval_runtime": 7.9947,
|
| 262 |
+
"eval_samples_per_second": 726.232,
|
| 263 |
+
"eval_steps_per_second": 45.405,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.76,
|
| 268 |
+
"learning_rate": 4.1e-05,
|
| 269 |
+
"loss": 0.5993,
|
| 270 |
+
"step": 9000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.76,
|
| 274 |
+
"eval_accuracy": 0.8645569620253165,
|
| 275 |
+
"eval_loss": 0.6027613282203674,
|
| 276 |
+
"eval_runtime": 8.0195,
|
| 277 |
+
"eval_samples_per_second": 723.989,
|
| 278 |
+
"eval_steps_per_second": 45.265,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.91,
|
| 283 |
+
"learning_rate": 4.05e-05,
|
| 284 |
+
"loss": 0.6527,
|
| 285 |
+
"step": 9500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.91,
|
| 289 |
+
"eval_accuracy": 0.8510214250124564,
|
| 290 |
+
"eval_loss": 0.6583752632141113,
|
| 291 |
+
"eval_runtime": 7.9195,
|
| 292 |
+
"eval_samples_per_second": 733.131,
|
| 293 |
+
"eval_steps_per_second": 45.836,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.06,
|
| 298 |
+
"learning_rate": 4e-05,
|
| 299 |
+
"loss": 0.5897,
|
| 300 |
+
"step": 10000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.06,
|
| 304 |
+
"eval_accuracy": 0.8676120587068623,
|
| 305 |
+
"eval_loss": 0.5727556943893433,
|
| 306 |
+
"eval_runtime": 7.9746,
|
| 307 |
+
"eval_samples_per_second": 728.065,
|
| 308 |
+
"eval_steps_per_second": 45.52,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.21,
|
| 313 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 314 |
+
"loss": 0.574,
|
| 315 |
+
"step": 10500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.21,
|
| 319 |
+
"eval_accuracy": 0.8670824400701618,
|
| 320 |
+
"eval_loss": 0.5869864821434021,
|
| 321 |
+
"eval_runtime": 7.8716,
|
| 322 |
+
"eval_samples_per_second": 737.59,
|
| 323 |
+
"eval_steps_per_second": 46.115,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.37,
|
| 328 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 329 |
+
"loss": 0.6026,
|
| 330 |
+
"step": 11000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.37,
|
| 334 |
+
"eval_accuracy": 0.8676513458361675,
|
| 335 |
+
"eval_loss": 0.6066599488258362,
|
| 336 |
+
"eval_runtime": 7.8242,
|
| 337 |
+
"eval_samples_per_second": 742.057,
|
| 338 |
+
"eval_steps_per_second": 46.395,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.52,
|
| 343 |
+
"learning_rate": 3.85e-05,
|
| 344 |
+
"loss": 0.5896,
|
| 345 |
+
"step": 11500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.52,
|
| 349 |
+
"eval_accuracy": 0.8638327806250629,
|
| 350 |
+
"eval_loss": 0.6000019311904907,
|
| 351 |
+
"eval_runtime": 8.0139,
|
| 352 |
+
"eval_samples_per_second": 724.49,
|
| 353 |
+
"eval_steps_per_second": 45.296,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 3.67,
|
| 358 |
+
"learning_rate": 3.8e-05,
|
| 359 |
+
"loss": 0.566,
|
| 360 |
+
"step": 12000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.67,
|
| 364 |
+
"eval_accuracy": 0.8711821948164563,
|
| 365 |
+
"eval_loss": 0.5566375851631165,
|
| 366 |
+
"eval_runtime": 7.8868,
|
| 367 |
+
"eval_samples_per_second": 736.17,
|
| 368 |
+
"eval_steps_per_second": 46.026,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.83,
|
| 373 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 374 |
+
"loss": 0.5928,
|
| 375 |
+
"step": 12500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.83,
|
| 379 |
+
"eval_accuracy": 0.8675352877307275,
|
| 380 |
+
"eval_loss": 0.5621004700660706,
|
| 381 |
+
"eval_runtime": 7.9912,
|
| 382 |
+
"eval_samples_per_second": 726.553,
|
| 383 |
+
"eval_steps_per_second": 45.425,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.98,
|
| 388 |
+
"learning_rate": 3.7e-05,
|
| 389 |
+
"loss": 0.597,
|
| 390 |
+
"step": 13000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 3.98,
|
| 394 |
+
"eval_accuracy": 0.8771320904403015,
|
| 395 |
+
"eval_loss": 0.5161893963813782,
|
| 396 |
+
"eval_runtime": 7.9771,
|
| 397 |
+
"eval_samples_per_second": 727.833,
|
| 398 |
+
"eval_steps_per_second": 45.505,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.13,
|
| 403 |
+
"learning_rate": 3.65e-05,
|
| 404 |
+
"loss": 0.5836,
|
| 405 |
+
"step": 13500
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.13,
|
| 409 |
+
"eval_accuracy": 0.8696463654223968,
|
| 410 |
+
"eval_loss": 0.5498046278953552,
|
| 411 |
+
"eval_runtime": 7.8463,
|
| 412 |
+
"eval_samples_per_second": 739.966,
|
| 413 |
+
"eval_steps_per_second": 46.264,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.29,
|
| 418 |
+
"learning_rate": 3.6e-05,
|
| 419 |
+
"loss": 0.5864,
|
| 420 |
+
"step": 14000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.29,
|
| 424 |
+
"eval_accuracy": 0.8639773945240183,
|
| 425 |
+
"eval_loss": 0.5728442072868347,
|
| 426 |
+
"eval_runtime": 7.8404,
|
| 427 |
+
"eval_samples_per_second": 740.524,
|
| 428 |
+
"eval_steps_per_second": 46.299,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.44,
|
| 433 |
+
"learning_rate": 3.55e-05,
|
| 434 |
+
"loss": 0.5562,
|
| 435 |
+
"step": 14500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.44,
|
| 439 |
+
"eval_accuracy": 0.8623497479643273,
|
| 440 |
+
"eval_loss": 0.6000498533248901,
|
| 441 |
+
"eval_runtime": 7.8135,
|
| 442 |
+
"eval_samples_per_second": 743.077,
|
| 443 |
+
"eval_steps_per_second": 46.458,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.59,
|
| 448 |
+
"learning_rate": 3.5e-05,
|
| 449 |
+
"loss": 0.5999,
|
| 450 |
+
"step": 15000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.59,
|
| 454 |
+
"eval_accuracy": 0.8679152291769344,
|
| 455 |
+
"eval_loss": 0.5589025020599365,
|
| 456 |
+
"eval_runtime": 7.7959,
|
| 457 |
+
"eval_samples_per_second": 744.749,
|
| 458 |
+
"eval_steps_per_second": 46.563,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.75,
|
| 463 |
+
"learning_rate": 3.45e-05,
|
| 464 |
+
"loss": 0.5767,
|
| 465 |
+
"step": 15500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.75,
|
| 469 |
+
"eval_accuracy": 0.8680821783151479,
|
| 470 |
+
"eval_loss": 0.5713112354278564,
|
| 471 |
+
"eval_runtime": 8.9874,
|
| 472 |
+
"eval_samples_per_second": 646.014,
|
| 473 |
+
"eval_steps_per_second": 40.39,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 4.9,
|
| 478 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 479 |
+
"loss": 0.5574,
|
| 480 |
+
"step": 16000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 4.9,
|
| 484 |
+
"eval_accuracy": 0.8739122026687295,
|
| 485 |
+
"eval_loss": 0.5337920784950256,
|
| 486 |
+
"eval_runtime": 10.8383,
|
| 487 |
+
"eval_samples_per_second": 535.691,
|
| 488 |
+
"eval_steps_per_second": 33.492,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 5.05,
|
| 493 |
+
"learning_rate": 3.35e-05,
|
| 494 |
+
"loss": 0.568,
|
| 495 |
+
"step": 16500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.05,
|
| 499 |
+
"eval_accuracy": 0.87250098000784,
|
| 500 |
+
"eval_loss": 0.552727222442627,
|
| 501 |
+
"eval_runtime": 7.8124,
|
| 502 |
+
"eval_samples_per_second": 743.18,
|
| 503 |
+
"eval_steps_per_second": 46.465,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 5.21,
|
| 508 |
+
"learning_rate": 3.3e-05,
|
| 509 |
+
"loss": 0.5568,
|
| 510 |
+
"step": 17000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 5.21,
|
| 514 |
+
"eval_accuracy": 0.8776927722971612,
|
| 515 |
+
"eval_loss": 0.5058096051216125,
|
| 516 |
+
"eval_runtime": 7.8143,
|
| 517 |
+
"eval_samples_per_second": 742.993,
|
| 518 |
+
"eval_steps_per_second": 46.453,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.36,
|
| 523 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 524 |
+
"loss": 0.5369,
|
| 525 |
+
"step": 17500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.36,
|
| 529 |
+
"eval_accuracy": 0.8719769673704415,
|
| 530 |
+
"eval_loss": 0.5599194169044495,
|
| 531 |
+
"eval_runtime": 7.8287,
|
| 532 |
+
"eval_samples_per_second": 741.628,
|
| 533 |
+
"eval_steps_per_second": 46.368,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 5.51,
|
| 538 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 539 |
+
"loss": 0.518,
|
| 540 |
+
"step": 18000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.51,
|
| 544 |
+
"eval_accuracy": 0.8720388349514563,
|
| 545 |
+
"eval_loss": 0.561033308506012,
|
| 546 |
+
"eval_runtime": 7.8241,
|
| 547 |
+
"eval_samples_per_second": 742.071,
|
| 548 |
+
"eval_steps_per_second": 46.395,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.66,
|
| 553 |
+
"learning_rate": 3.15e-05,
|
| 554 |
+
"loss": 0.5637,
|
| 555 |
+
"step": 18500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.66,
|
| 559 |
+
"eval_accuracy": 0.8727518855153742,
|
| 560 |
+
"eval_loss": 0.5467284917831421,
|
| 561 |
+
"eval_runtime": 8.0155,
|
| 562 |
+
"eval_samples_per_second": 724.347,
|
| 563 |
+
"eval_steps_per_second": 45.287,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 5.82,
|
| 568 |
+
"learning_rate": 3.1e-05,
|
| 569 |
+
"loss": 0.557,
|
| 570 |
+
"step": 19000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 5.82,
|
| 574 |
+
"eval_accuracy": 0.8713813872158539,
|
| 575 |
+
"eval_loss": 0.5348953604698181,
|
| 576 |
+
"eval_runtime": 8.0121,
|
| 577 |
+
"eval_samples_per_second": 724.653,
|
| 578 |
+
"eval_steps_per_second": 45.306,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.97,
|
| 583 |
+
"learning_rate": 3.05e-05,
|
| 584 |
+
"loss": 0.5499,
|
| 585 |
+
"step": 19500
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.97,
|
| 589 |
+
"eval_accuracy": 0.8724001160878398,
|
| 590 |
+
"eval_loss": 0.5467893481254578,
|
| 591 |
+
"eval_runtime": 7.7511,
|
| 592 |
+
"eval_samples_per_second": 749.05,
|
| 593 |
+
"eval_steps_per_second": 46.832,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 6.12,
|
| 598 |
+
"learning_rate": 3e-05,
|
| 599 |
+
"loss": 0.5304,
|
| 600 |
+
"step": 20000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 6.12,
|
| 604 |
+
"eval_accuracy": 0.8740521910388971,
|
| 605 |
+
"eval_loss": 0.5243064761161804,
|
| 606 |
+
"eval_runtime": 7.8201,
|
| 607 |
+
"eval_samples_per_second": 742.45,
|
| 608 |
+
"eval_steps_per_second": 46.419,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.28,
|
| 613 |
+
"learning_rate": 2.95e-05,
|
| 614 |
+
"loss": 0.5431,
|
| 615 |
+
"step": 20500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.28,
|
| 619 |
+
"eval_accuracy": 0.8783942176206291,
|
| 620 |
+
"eval_loss": 0.4997641146183014,
|
| 621 |
+
"eval_runtime": 8.0018,
|
| 622 |
+
"eval_samples_per_second": 725.585,
|
| 623 |
+
"eval_steps_per_second": 45.365,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.43,
|
| 628 |
+
"learning_rate": 2.9e-05,
|
| 629 |
+
"loss": 0.5508,
|
| 630 |
+
"step": 21000
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.43,
|
| 634 |
+
"eval_accuracy": 0.8763812154696132,
|
| 635 |
+
"eval_loss": 0.5366745591163635,
|
| 636 |
+
"eval_runtime": 7.8074,
|
| 637 |
+
"eval_samples_per_second": 743.654,
|
| 638 |
+
"eval_steps_per_second": 46.494,
|
| 639 |
+
"step": 21000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.58,
|
| 643 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 644 |
+
"loss": 0.5701,
|
| 645 |
+
"step": 21500
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.58,
|
| 649 |
+
"eval_accuracy": 0.8734250823803063,
|
| 650 |
+
"eval_loss": 0.5364522337913513,
|
| 651 |
+
"eval_runtime": 7.9868,
|
| 652 |
+
"eval_samples_per_second": 726.947,
|
| 653 |
+
"eval_steps_per_second": 45.45,
|
| 654 |
+
"step": 21500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 6.74,
|
| 658 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 659 |
+
"loss": 0.521,
|
| 660 |
+
"step": 22000
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 6.74,
|
| 664 |
+
"eval_accuracy": 0.8818635607321131,
|
| 665 |
+
"eval_loss": 0.4879148006439209,
|
| 666 |
+
"eval_runtime": 7.9938,
|
| 667 |
+
"eval_samples_per_second": 726.31,
|
| 668 |
+
"eval_steps_per_second": 45.41,
|
| 669 |
+
"step": 22000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 6.89,
|
| 673 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 674 |
+
"loss": 0.5514,
|
| 675 |
+
"step": 22500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 6.89,
|
| 679 |
+
"eval_accuracy": 0.8786950074147306,
|
| 680 |
+
"eval_loss": 0.5105842351913452,
|
| 681 |
+
"eval_runtime": 7.8325,
|
| 682 |
+
"eval_samples_per_second": 741.269,
|
| 683 |
+
"eval_steps_per_second": 46.345,
|
| 684 |
+
"step": 22500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 7.04,
|
| 688 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 689 |
+
"loss": 0.547,
|
| 690 |
+
"step": 23000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 7.04,
|
| 694 |
+
"eval_accuracy": 0.8747058823529412,
|
| 695 |
+
"eval_loss": 0.5258113741874695,
|
| 696 |
+
"eval_runtime": 7.8237,
|
| 697 |
+
"eval_samples_per_second": 742.1,
|
| 698 |
+
"eval_steps_per_second": 46.397,
|
| 699 |
+
"step": 23000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 7.2,
|
| 703 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 704 |
+
"loss": 0.5512,
|
| 705 |
+
"step": 23500
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.2,
|
| 709 |
+
"eval_accuracy": 0.877830692973078,
|
| 710 |
+
"eval_loss": 0.49750423431396484,
|
| 711 |
+
"eval_runtime": 7.9086,
|
| 712 |
+
"eval_samples_per_second": 734.135,
|
| 713 |
+
"eval_steps_per_second": 45.899,
|
| 714 |
+
"step": 23500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 7.35,
|
| 718 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 719 |
+
"loss": 0.5407,
|
| 720 |
+
"step": 24000
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 7.35,
|
| 724 |
+
"eval_accuracy": 0.8785601265822784,
|
| 725 |
+
"eval_loss": 0.494391530752182,
|
| 726 |
+
"eval_runtime": 8.2168,
|
| 727 |
+
"eval_samples_per_second": 706.599,
|
| 728 |
+
"eval_steps_per_second": 44.178,
|
| 729 |
+
"step": 24000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.5,
|
| 733 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 734 |
+
"loss": 0.5181,
|
| 735 |
+
"step": 24500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 7.5,
|
| 739 |
+
"eval_accuracy": 0.8794734275962945,
|
| 740 |
+
"eval_loss": 0.4911736845970154,
|
| 741 |
+
"eval_runtime": 8.2044,
|
| 742 |
+
"eval_samples_per_second": 707.673,
|
| 743 |
+
"eval_steps_per_second": 44.245,
|
| 744 |
+
"step": 24500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 7.65,
|
| 748 |
+
"learning_rate": 2.5e-05,
|
| 749 |
+
"loss": 0.5493,
|
| 750 |
+
"step": 25000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 7.65,
|
| 754 |
+
"eval_accuracy": 0.87302207462395,
|
| 755 |
+
"eval_loss": 0.5187950730323792,
|
| 756 |
+
"eval_runtime": 8.0486,
|
| 757 |
+
"eval_samples_per_second": 721.366,
|
| 758 |
+
"eval_steps_per_second": 45.101,
|
| 759 |
+
"step": 25000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.81,
|
| 763 |
+
"learning_rate": 2.45e-05,
|
| 764 |
+
"loss": 0.5388,
|
| 765 |
+
"step": 25500
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 7.81,
|
| 769 |
+
"eval_accuracy": 0.8831105473704751,
|
| 770 |
+
"eval_loss": 0.5000073313713074,
|
| 771 |
+
"eval_runtime": 8.0362,
|
| 772 |
+
"eval_samples_per_second": 722.481,
|
| 773 |
+
"eval_steps_per_second": 45.171,
|
| 774 |
+
"step": 25500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 7.96,
|
| 778 |
+
"learning_rate": 2.4e-05,
|
| 779 |
+
"loss": 0.5284,
|
| 780 |
+
"step": 26000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 7.96,
|
| 784 |
+
"eval_accuracy": 0.8737309019221291,
|
| 785 |
+
"eval_loss": 0.5161213278770447,
|
| 786 |
+
"eval_runtime": 8.1271,
|
| 787 |
+
"eval_samples_per_second": 714.401,
|
| 788 |
+
"eval_steps_per_second": 44.665,
|
| 789 |
+
"step": 26000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 8.11,
|
| 793 |
+
"learning_rate": 2.35e-05,
|
| 794 |
+
"loss": 0.5116,
|
| 795 |
+
"step": 26500
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.11,
|
| 799 |
+
"eval_accuracy": 0.8759842519685039,
|
| 800 |
+
"eval_loss": 0.5262829065322876,
|
| 801 |
+
"eval_runtime": 8.1731,
|
| 802 |
+
"eval_samples_per_second": 710.381,
|
| 803 |
+
"eval_steps_per_second": 44.414,
|
| 804 |
+
"step": 26500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 8.27,
|
| 808 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 809 |
+
"loss": 0.5161,
|
| 810 |
+
"step": 27000
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 8.27,
|
| 814 |
+
"eval_accuracy": 0.8786888397577621,
|
| 815 |
+
"eval_loss": 0.500228762626648,
|
| 816 |
+
"eval_runtime": 8.034,
|
| 817 |
+
"eval_samples_per_second": 722.681,
|
| 818 |
+
"eval_steps_per_second": 45.183,
|
| 819 |
+
"step": 27000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 8.42,
|
| 823 |
+
"learning_rate": 2.25e-05,
|
| 824 |
+
"loss": 0.5185,
|
| 825 |
+
"step": 27500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.42,
|
| 829 |
+
"eval_accuracy": 0.8744550138723741,
|
| 830 |
+
"eval_loss": 0.5127227902412415,
|
| 831 |
+
"eval_runtime": 7.7182,
|
| 832 |
+
"eval_samples_per_second": 752.252,
|
| 833 |
+
"eval_steps_per_second": 47.032,
|
| 834 |
+
"step": 27500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 8.57,
|
| 838 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 839 |
+
"loss": 0.5291,
|
| 840 |
+
"step": 28000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 8.57,
|
| 844 |
+
"eval_accuracy": 0.8782496527088708,
|
| 845 |
+
"eval_loss": 0.5115563273429871,
|
| 846 |
+
"eval_runtime": 8.0802,
|
| 847 |
+
"eval_samples_per_second": 718.543,
|
| 848 |
+
"eval_steps_per_second": 44.924,
|
| 849 |
+
"step": 28000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 8.73,
|
| 853 |
+
"learning_rate": 2.15e-05,
|
| 854 |
+
"loss": 0.5061,
|
| 855 |
+
"step": 28500
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 8.73,
|
| 859 |
+
"eval_accuracy": 0.8773942634905202,
|
| 860 |
+
"eval_loss": 0.4972003996372223,
|
| 861 |
+
"eval_runtime": 7.7937,
|
| 862 |
+
"eval_samples_per_second": 744.959,
|
| 863 |
+
"eval_steps_per_second": 46.576,
|
| 864 |
+
"step": 28500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 8.88,
|
| 868 |
+
"learning_rate": 2.1e-05,
|
| 869 |
+
"loss": 0.479,
|
| 870 |
+
"step": 29000
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 8.88,
|
| 874 |
+
"eval_accuracy": 0.8797814207650273,
|
| 875 |
+
"eval_loss": 0.49780747294425964,
|
| 876 |
+
"eval_runtime": 7.8838,
|
| 877 |
+
"eval_samples_per_second": 736.449,
|
| 878 |
+
"eval_steps_per_second": 46.044,
|
| 879 |
+
"step": 29000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 9.03,
|
| 883 |
+
"learning_rate": 2.05e-05,
|
| 884 |
+
"loss": 0.5154,
|
| 885 |
+
"step": 29500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 9.03,
|
| 889 |
+
"eval_accuracy": 0.877119904790241,
|
| 890 |
+
"eval_loss": 0.5088150501251221,
|
| 891 |
+
"eval_runtime": 7.7019,
|
| 892 |
+
"eval_samples_per_second": 753.843,
|
| 893 |
+
"eval_steps_per_second": 47.131,
|
| 894 |
+
"step": 29500
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 9.19,
|
| 898 |
+
"learning_rate": 2e-05,
|
| 899 |
+
"loss": 0.4989,
|
| 900 |
+
"step": 30000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 9.19,
|
| 904 |
+
"eval_accuracy": 0.8744332741967278,
|
| 905 |
+
"eval_loss": 0.5118668079376221,
|
| 906 |
+
"eval_runtime": 7.7316,
|
| 907 |
+
"eval_samples_per_second": 750.942,
|
| 908 |
+
"eval_steps_per_second": 46.95,
|
| 909 |
+
"step": 30000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 9.34,
|
| 913 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 914 |
+
"loss": 0.5098,
|
| 915 |
+
"step": 30500
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 9.34,
|
| 919 |
+
"eval_accuracy": 0.8825979176995538,
|
| 920 |
+
"eval_loss": 0.4915599524974823,
|
| 921 |
+
"eval_runtime": 7.7036,
|
| 922 |
+
"eval_samples_per_second": 753.676,
|
| 923 |
+
"eval_steps_per_second": 47.121,
|
| 924 |
+
"step": 30500
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 9.49,
|
| 928 |
+
"learning_rate": 1.9e-05,
|
| 929 |
+
"loss": 0.4777,
|
| 930 |
+
"step": 31000
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 9.49,
|
| 934 |
+
"eval_accuracy": 0.882445449184195,
|
| 935 |
+
"eval_loss": 0.49568039178848267,
|
| 936 |
+
"eval_runtime": 7.7025,
|
| 937 |
+
"eval_samples_per_second": 753.779,
|
| 938 |
+
"eval_steps_per_second": 47.127,
|
| 939 |
+
"step": 31000
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 9.64,
|
| 943 |
+
"learning_rate": 1.85e-05,
|
| 944 |
+
"loss": 0.5462,
|
| 945 |
+
"step": 31500
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 9.64,
|
| 949 |
+
"eval_accuracy": 0.8778625954198473,
|
| 950 |
+
"eval_loss": 0.48457399010658264,
|
| 951 |
+
"eval_runtime": 7.7115,
|
| 952 |
+
"eval_samples_per_second": 752.903,
|
| 953 |
+
"eval_steps_per_second": 47.073,
|
| 954 |
+
"step": 31500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 9.8,
|
| 958 |
+
"learning_rate": 1.8e-05,
|
| 959 |
+
"loss": 0.509,
|
| 960 |
+
"step": 32000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 9.8,
|
| 964 |
+
"eval_accuracy": 0.8810146190337884,
|
| 965 |
+
"eval_loss": 0.48734790086746216,
|
| 966 |
+
"eval_runtime": 7.7302,
|
| 967 |
+
"eval_samples_per_second": 751.078,
|
| 968 |
+
"eval_steps_per_second": 46.959,
|
| 969 |
+
"step": 32000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 9.95,
|
| 973 |
+
"learning_rate": 1.75e-05,
|
| 974 |
+
"loss": 0.5181,
|
| 975 |
+
"step": 32500
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 9.95,
|
| 979 |
+
"eval_accuracy": 0.8710217755443886,
|
| 980 |
+
"eval_loss": 0.5227355360984802,
|
| 981 |
+
"eval_runtime": 7.7073,
|
| 982 |
+
"eval_samples_per_second": 753.31,
|
| 983 |
+
"eval_steps_per_second": 47.098,
|
| 984 |
+
"step": 32500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 10.1,
|
| 988 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 989 |
+
"loss": 0.5269,
|
| 990 |
+
"step": 33000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 10.1,
|
| 994 |
+
"eval_accuracy": 0.8802636757182212,
|
| 995 |
+
"eval_loss": 0.49287834763526917,
|
| 996 |
+
"eval_runtime": 8.3473,
|
| 997 |
+
"eval_samples_per_second": 695.551,
|
| 998 |
+
"eval_steps_per_second": 43.487,
|
| 999 |
+
"step": 33000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 10.26,
|
| 1003 |
+
"learning_rate": 1.65e-05,
|
| 1004 |
+
"loss": 0.5094,
|
| 1005 |
+
"step": 33500
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 10.26,
|
| 1009 |
+
"eval_accuracy": 0.8877481840193705,
|
| 1010 |
+
"eval_loss": 0.4840761125087738,
|
| 1011 |
+
"eval_runtime": 8.4693,
|
| 1012 |
+
"eval_samples_per_second": 685.535,
|
| 1013 |
+
"eval_steps_per_second": 42.861,
|
| 1014 |
+
"step": 33500
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 10.41,
|
| 1018 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 1019 |
+
"loss": 0.5033,
|
| 1020 |
+
"step": 34000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 10.41,
|
| 1024 |
+
"eval_accuracy": 0.8805490654205608,
|
| 1025 |
+
"eval_loss": 0.5128547549247742,
|
| 1026 |
+
"eval_runtime": 8.1006,
|
| 1027 |
+
"eval_samples_per_second": 716.736,
|
| 1028 |
+
"eval_steps_per_second": 44.811,
|
| 1029 |
+
"step": 34000
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 10.56,
|
| 1033 |
+
"learning_rate": 1.55e-05,
|
| 1034 |
+
"loss": 0.4913,
|
| 1035 |
+
"step": 34500
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 10.56,
|
| 1039 |
+
"eval_accuracy": 0.8789432939810334,
|
| 1040 |
+
"eval_loss": 0.4978225529193878,
|
| 1041 |
+
"eval_runtime": 8.4845,
|
| 1042 |
+
"eval_samples_per_second": 684.304,
|
| 1043 |
+
"eval_steps_per_second": 42.784,
|
| 1044 |
+
"step": 34500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 10.72,
|
| 1048 |
+
"learning_rate": 1.5e-05,
|
| 1049 |
+
"loss": 0.4938,
|
| 1050 |
+
"step": 35000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 10.72,
|
| 1054 |
+
"eval_accuracy": 0.8838202465301368,
|
| 1055 |
+
"eval_loss": 0.46402791142463684,
|
| 1056 |
+
"eval_runtime": 8.2894,
|
| 1057 |
+
"eval_samples_per_second": 700.417,
|
| 1058 |
+
"eval_steps_per_second": 43.791,
|
| 1059 |
+
"step": 35000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 10.87,
|
| 1063 |
+
"learning_rate": 1.45e-05,
|
| 1064 |
+
"loss": 0.4954,
|
| 1065 |
+
"step": 35500
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 10.87,
|
| 1069 |
+
"eval_accuracy": 0.8793576184880533,
|
| 1070 |
+
"eval_loss": 0.4990694522857666,
|
| 1071 |
+
"eval_runtime": 8.2824,
|
| 1072 |
+
"eval_samples_per_second": 701.003,
|
| 1073 |
+
"eval_steps_per_second": 43.828,
|
| 1074 |
+
"step": 35500
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 11.02,
|
| 1078 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 1079 |
+
"loss": 0.458,
|
| 1080 |
+
"step": 36000
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 11.02,
|
| 1084 |
+
"eval_accuracy": 0.8885711468297012,
|
| 1085 |
+
"eval_loss": 0.4452793300151825,
|
| 1086 |
+
"eval_runtime": 8.4212,
|
| 1087 |
+
"eval_samples_per_second": 689.451,
|
| 1088 |
+
"eval_steps_per_second": 43.106,
|
| 1089 |
+
"step": 36000
|
| 1090 |
+
}
|
| 1091 |
+
],
|
| 1092 |
+
"logging_steps": 500,
|
| 1093 |
+
"max_steps": 50000,
|
| 1094 |
+
"num_train_epochs": 16,
|
| 1095 |
+
"save_steps": 500,
|
| 1096 |
+
"total_flos": 5426535158775808.0,
|
| 1097 |
+
"trial_name": null,
|
| 1098 |
+
"trial_params": null
|
| 1099 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11847a8f7a409c8f6722c932668136e0c8a7072d934ba9c10d1f2aff07bd3ba4
|
| 3 |
+
size 4091
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|