Joeran Bosma
commited on
Commit
·
e44fd8f
1
Parent(s):
938852c
Initial release
Browse files- README.md +104 -3
- all_results.json +15 -0
- config.json +26 -0
- eval_results.json +10 -0
- generation_config.json +5 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +15 -0
- train_results.json +8 -0
- trainer_state.json +3046 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
CHANGED
|
@@ -1,3 +1,104 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-nc-sa-4.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-sa-4.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
# DRAGON BERT base mixed-domain
|
| 6 |
+
|
| 7 |
+
Pretrained model on Dutch clinical reports using a masked language modeling (MLM) objective. It was introduced in [this](#pending) paper. The model was first pretrained using general domain data, as specified [here](https://huggingface.co/bert-base-multilingual-cased). The pretrained model was taken from HuggingFace: [`bert-base-multilingual-cased`](https://huggingface.co/bert-base-multilingual-cased). Subsequently, the model was pretrained using domain-specific data (i.e., clinical reports). The tokenizer of [`bert-base-multilingual-cased`](https://huggingface.co/bert-base-multilingual-cased) was used.
|
| 8 |
+
|
| 9 |
+
## Model description
|
| 10 |
+
BERT is a transformers model that was pretrained on a large corpus of Dutch clinical reports in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labeling them in any way with an automatic process to generate inputs and labels from those texts.
|
| 11 |
+
|
| 12 |
+
This way, the model learns an inner representation of the Dutch medical language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled reports, for instance, you can train a standard classifier using the features produced by the BERT model as inputs.
|
| 13 |
+
|
| 14 |
+
## Model variations
|
| 15 |
+
Multiple architectures were pretrained for the DRAGON challenge.
|
| 16 |
+
|
| 17 |
+
| Model | #params | Language |
|
| 18 |
+
|------------------------|--------------------------------|-------|
|
| 19 |
+
| [`joeranbosma/dragon-bert-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-bert-base-mixed-domain) | 109M | Dutch → Dutch |
|
| 20 |
+
| [`joeranbosma/dragon-roberta-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-base-mixed-domain) | 278M | Multiple → Dutch |
|
| 21 |
+
| [`joeranbosma/dragon-roberta-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-large-mixed-domain) | 560M | Multiple → Dutch |
|
| 22 |
+
| [`joeranbosma/dragon-longformer-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-base-mixed-domain) | 149M | English → Dutch |
|
| 23 |
+
| [`joeranbosma/dragon-longformer-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-large-mixed-domain) | 435M | English → Dutch |
|
| 24 |
+
| [`joeranbosma/dragon-bert-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-bert-base-domain-specific) | 109M | Dutch |
|
| 25 |
+
| [`joeranbosma/dragon-roberta-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-base-domain-specific) | 278M | Dutch |
|
| 26 |
+
| [`joeranbosma/dragon-roberta-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-large-domain-specific) | 560M | Dutch |
|
| 27 |
+
| [`joeranbosma/dragon-longformer-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-base-domain-specific) | 149M | Dutch |
|
| 28 |
+
| [`joeranbosma/dragon-longformer-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-large-domain-specific) | 435M | Dutch |
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
## Intended uses & limitations
|
| 32 |
+
You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task.
|
| 33 |
+
|
| 34 |
+
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole text (e.g., a clinical report) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2.
|
| 35 |
+
|
| 36 |
+
## How to use
|
| 37 |
+
You can use this model directly with a pipeline for masked language modeling:
|
| 38 |
+
|
| 39 |
+
```python
|
| 40 |
+
from transformers import pipeline
|
| 41 |
+
unmasker = pipeline("fill-mask", model="joeranbosma/dragon-bert-base-mixed-domain")
|
| 42 |
+
unmasker("Dit onderzoek geen aanwijzingen voor significant carcinoom. PIRADS <mask>.")
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
Here is how to use this model to get the features of a given text in PyTorch:
|
| 46 |
+
|
| 47 |
+
```python
|
| 48 |
+
from transformers import AutoTokenizer, AutoModel
|
| 49 |
+
tokenizer = AutoTokenizer.from_pretrained("joeranbosma/dragon-bert-base-mixed-domain")
|
| 50 |
+
model = AutoModel.from_pretrained("joeranbosma/dragon-bert-base-mixed-domain")
|
| 51 |
+
text = "Replace me by any text you'd like."
|
| 52 |
+
encoded_input = tokenizer(text, return_tensors="pt")
|
| 53 |
+
output = model(**encoded_input)
|
| 54 |
+
```
|
| 55 |
+
|
| 56 |
+
## Limitations and bias
|
| 57 |
+
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model.
|
| 58 |
+
|
| 59 |
+
## Training data
|
| 60 |
+
For pretraining, 4,333,201 clinical reports (466,351 consecutive patients) were selected from Ziekenhuisgroep Twente from patients with a diagnostic or interventional visit between 13 July 2000 and 25 April 2023. 180,439 duplicate clinical reports (179,808 patients) were excluded, resulting in 4,152,762 included reports (463,692 patients). These reports were split into training (80%, 3,322,209 reports), validation (10%, 415,276 reports), and testing (10%, 415,277 reports). The testing reports were set aside for future analysis and are not used for pretraining.
|
| 61 |
+
|
| 62 |
+
## Training procedure
|
| 63 |
+
|
| 64 |
+
### Pretraining
|
| 65 |
+
The model was pretrained using masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then runs the entire masked sentence through the model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally masks the future tokens. It allows the model to learn a bidirectional representation of the sentence.
|
| 66 |
+
|
| 67 |
+
The details of the masking procedure for each sentence are the following:
|
| 68 |
+
- 15% of the tokens are masked.
|
| 69 |
+
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
|
| 70 |
+
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
|
| 71 |
+
- In the 10% remaining cases, the masked tokens are left as is.
|
| 72 |
+
|
| 73 |
+
The HuggingFace implementation was used for pretraining: [`run_mlm.py`](https://github.com/huggingface/transformers/blob/7c6ec195adbfcd22cb6baeee64dd3c24a4b80c74/examples/pytorch/language-modeling/run_mlm.py).
|
| 74 |
+
|
| 75 |
+
### Pretraining hyperparameters
|
| 76 |
+
|
| 77 |
+
The following hyperparameters were used during pretraining:
|
| 78 |
+
- `learning_rate`: 5e-05
|
| 79 |
+
- `train_batch_size`: 4
|
| 80 |
+
- `eval_batch_size`: 4
|
| 81 |
+
- `seed`: 42
|
| 82 |
+
- `gradient_accumulation_steps`: 4
|
| 83 |
+
- `total_train_batch_size`: 16
|
| 84 |
+
- `optimizer`: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 85 |
+
- `lr_scheduler_type`: linear
|
| 86 |
+
- `num_epochs`: 3.0
|
| 87 |
+
- `max_seq_length`: 512
|
| 88 |
+
|
| 89 |
+
### Framework versions
|
| 90 |
+
|
| 91 |
+
- Transformers 4.29.0.dev0
|
| 92 |
+
- Pytorch 2.0.0+cu117
|
| 93 |
+
- Datasets 2.11.0
|
| 94 |
+
- Tokenizers 0.13.3
|
| 95 |
+
|
| 96 |
+
## Evaluation results
|
| 97 |
+
|
| 98 |
+
Pending evaluation on the DRAGON benchmark.
|
| 99 |
+
|
| 100 |
+
### BibTeX entry and citation info
|
| 101 |
+
|
| 102 |
+
```bibtex
|
| 103 |
+
@article{PENDING}
|
| 104 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9075239480049601,
|
| 4 |
+
"eval_loss": 0.37417328357696533,
|
| 5 |
+
"eval_runtime": 2158.2128,
|
| 6 |
+
"eval_samples": 146663,
|
| 7 |
+
"eval_samples_per_second": 67.956,
|
| 8 |
+
"eval_steps_per_second": 16.989,
|
| 9 |
+
"perplexity": 1.4537890463976202,
|
| 10 |
+
"train_loss": 0.5021202932867165,
|
| 11 |
+
"train_runtime": 238578.5378,
|
| 12 |
+
"train_samples": 1171723,
|
| 13 |
+
"train_samples_per_second": 14.734,
|
| 14 |
+
"train_steps_per_second": 0.921
|
| 15 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "GroNLP/bert-base-dutch-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 3,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.29.0.dev0",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30073
|
| 26 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9075239480049601,
|
| 4 |
+
"eval_loss": 0.37417328357696533,
|
| 5 |
+
"eval_runtime": 2158.2128,
|
| 6 |
+
"eval_samples": 146663,
|
| 7 |
+
"eval_samples_per_second": 67.956,
|
| 8 |
+
"eval_steps_per_second": 16.989,
|
| 9 |
+
"perplexity": 1.4537890463976202
|
| 10 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 3,
|
| 4 |
+
"transformers_version": "4.29.0.dev0"
|
| 5 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31dae0fa61b8995c203138b305cbf61c6905c2c84f00c6a26af91d497f2e53e8
|
| 3 |
+
size 436749497
|
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 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_basic_tokenize": true,
|
| 5 |
+
"do_lower_case": false,
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 512,
|
| 8 |
+
"never_split": null,
|
| 9 |
+
"pad_token": "[PAD]",
|
| 10 |
+
"sep_token": "[SEP]",
|
| 11 |
+
"strip_accents": null,
|
| 12 |
+
"tokenize_chinese_chars": true,
|
| 13 |
+
"tokenizer_class": "BertTokenizer",
|
| 14 |
+
"unk_token": "[UNK]"
|
| 15 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.5021202932867165,
|
| 4 |
+
"train_runtime": 238578.5378,
|
| 5 |
+
"train_samples": 1171723,
|
| 6 |
+
"train_samples_per_second": 14.734,
|
| 7 |
+
"train_steps_per_second": 0.921
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3046 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3740881383419037,
|
| 3 |
+
"best_model_checkpoint": "/output/bert-base-dutch-cased-mlm-zgt-radpat/checkpoint-215000",
|
| 4 |
+
"epoch": 2.9999692760411154,
|
| 5 |
+
"global_step": 219696,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 4.98862063942903e-05,
|
| 13 |
+
"loss": 2.2457,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 4.9772412788580583e-05,
|
| 19 |
+
"loss": 1.5618,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 4.965861918287088e-05,
|
| 25 |
+
"loss": 1.3494,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 4.954482557716117e-05,
|
| 31 |
+
"loss": 1.2085,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.03,
|
| 36 |
+
"learning_rate": 4.943103197145146e-05,
|
| 37 |
+
"loss": 1.1219,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04,
|
| 42 |
+
"learning_rate": 4.931723836574175e-05,
|
| 43 |
+
"loss": 1.0672,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"learning_rate": 4.920344476003205e-05,
|
| 49 |
+
"loss": 1.019,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.05,
|
| 54 |
+
"learning_rate": 4.908965115432234e-05,
|
| 55 |
+
"loss": 0.9802,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.06,
|
| 60 |
+
"learning_rate": 4.897585754861263e-05,
|
| 61 |
+
"loss": 0.9458,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.07,
|
| 66 |
+
"learning_rate": 4.886206394290293e-05,
|
| 67 |
+
"loss": 0.9108,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.07,
|
| 72 |
+
"eval_accuracy": 0.8237946784561654,
|
| 73 |
+
"eval_loss": 0.8260091543197632,
|
| 74 |
+
"eval_runtime": 2158.7476,
|
| 75 |
+
"eval_samples_per_second": 67.939,
|
| 76 |
+
"eval_steps_per_second": 16.985,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.08,
|
| 81 |
+
"learning_rate": 4.874827033719321e-05,
|
| 82 |
+
"loss": 0.8877,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.08,
|
| 87 |
+
"learning_rate": 4.863447673148351e-05,
|
| 88 |
+
"loss": 0.8687,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.09,
|
| 93 |
+
"learning_rate": 4.85206831257738e-05,
|
| 94 |
+
"loss": 0.8495,
|
| 95 |
+
"step": 6500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.1,
|
| 99 |
+
"learning_rate": 4.840688952006409e-05,
|
| 100 |
+
"loss": 0.8388,
|
| 101 |
+
"step": 7000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.1,
|
| 105 |
+
"learning_rate": 4.829309591435438e-05,
|
| 106 |
+
"loss": 0.8158,
|
| 107 |
+
"step": 7500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.817930230864468e-05,
|
| 112 |
+
"loss": 0.8081,
|
| 113 |
+
"step": 8000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.12,
|
| 117 |
+
"learning_rate": 4.806550870293497e-05,
|
| 118 |
+
"loss": 0.7942,
|
| 119 |
+
"step": 8500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.12,
|
| 123 |
+
"learning_rate": 4.795171509722526e-05,
|
| 124 |
+
"loss": 0.7903,
|
| 125 |
+
"step": 9000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.13,
|
| 129 |
+
"learning_rate": 4.7837921491515556e-05,
|
| 130 |
+
"loss": 0.7759,
|
| 131 |
+
"step": 9500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.14,
|
| 135 |
+
"learning_rate": 4.772412788580584e-05,
|
| 136 |
+
"loss": 0.7611,
|
| 137 |
+
"step": 10000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.14,
|
| 141 |
+
"eval_accuracy": 0.8465949226392128,
|
| 142 |
+
"eval_loss": 0.6964662671089172,
|
| 143 |
+
"eval_runtime": 2158.1384,
|
| 144 |
+
"eval_samples_per_second": 67.958,
|
| 145 |
+
"eval_steps_per_second": 16.99,
|
| 146 |
+
"step": 10000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.14,
|
| 150 |
+
"learning_rate": 4.761033428009614e-05,
|
| 151 |
+
"loss": 0.7551,
|
| 152 |
+
"step": 10500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.15,
|
| 156 |
+
"learning_rate": 4.749654067438643e-05,
|
| 157 |
+
"loss": 0.7481,
|
| 158 |
+
"step": 11000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.16,
|
| 162 |
+
"learning_rate": 4.738274706867672e-05,
|
| 163 |
+
"loss": 0.7398,
|
| 164 |
+
"step": 11500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.16,
|
| 168 |
+
"learning_rate": 4.7268953462967015e-05,
|
| 169 |
+
"loss": 0.7328,
|
| 170 |
+
"step": 12000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.17,
|
| 174 |
+
"learning_rate": 4.7155159857257306e-05,
|
| 175 |
+
"loss": 0.7163,
|
| 176 |
+
"step": 12500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.18,
|
| 180 |
+
"learning_rate": 4.7041366251547596e-05,
|
| 181 |
+
"loss": 0.712,
|
| 182 |
+
"step": 13000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.18,
|
| 186 |
+
"learning_rate": 4.692757264583789e-05,
|
| 187 |
+
"loss": 0.7099,
|
| 188 |
+
"step": 13500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.19,
|
| 192 |
+
"learning_rate": 4.681377904012818e-05,
|
| 193 |
+
"loss": 0.7059,
|
| 194 |
+
"step": 14000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.2,
|
| 198 |
+
"learning_rate": 4.669998543441847e-05,
|
| 199 |
+
"loss": 0.6952,
|
| 200 |
+
"step": 14500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.2,
|
| 204 |
+
"learning_rate": 4.6586191828708765e-05,
|
| 205 |
+
"loss": 0.6909,
|
| 206 |
+
"step": 15000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2,
|
| 210 |
+
"eval_accuracy": 0.8575000699065417,
|
| 211 |
+
"eval_loss": 0.6342362761497498,
|
| 212 |
+
"eval_runtime": 2158.6043,
|
| 213 |
+
"eval_samples_per_second": 67.943,
|
| 214 |
+
"eval_steps_per_second": 16.986,
|
| 215 |
+
"step": 15000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.21,
|
| 219 |
+
"learning_rate": 4.6472398222999056e-05,
|
| 220 |
+
"loss": 0.6902,
|
| 221 |
+
"step": 15500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.22,
|
| 225 |
+
"learning_rate": 4.6358604617289346e-05,
|
| 226 |
+
"loss": 0.6837,
|
| 227 |
+
"step": 16000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.23,
|
| 231 |
+
"learning_rate": 4.624481101157964e-05,
|
| 232 |
+
"loss": 0.674,
|
| 233 |
+
"step": 16500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.23,
|
| 237 |
+
"learning_rate": 4.6131017405869934e-05,
|
| 238 |
+
"loss": 0.6747,
|
| 239 |
+
"step": 17000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.24,
|
| 243 |
+
"learning_rate": 4.6017223800160224e-05,
|
| 244 |
+
"loss": 0.6669,
|
| 245 |
+
"step": 17500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.25,
|
| 249 |
+
"learning_rate": 4.5903430194450515e-05,
|
| 250 |
+
"loss": 0.6659,
|
| 251 |
+
"step": 18000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.25,
|
| 255 |
+
"learning_rate": 4.5789636588740805e-05,
|
| 256 |
+
"loss": 0.6612,
|
| 257 |
+
"step": 18500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.26,
|
| 261 |
+
"learning_rate": 4.5675842983031096e-05,
|
| 262 |
+
"loss": 0.6607,
|
| 263 |
+
"step": 19000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.27,
|
| 267 |
+
"learning_rate": 4.556204937732139e-05,
|
| 268 |
+
"loss": 0.6516,
|
| 269 |
+
"step": 19500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.27,
|
| 273 |
+
"learning_rate": 4.5448255771611684e-05,
|
| 274 |
+
"loss": 0.6476,
|
| 275 |
+
"step": 20000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.27,
|
| 279 |
+
"eval_accuracy": 0.8644562505743907,
|
| 280 |
+
"eval_loss": 0.5950246453285217,
|
| 281 |
+
"eval_runtime": 2156.4823,
|
| 282 |
+
"eval_samples_per_second": 68.01,
|
| 283 |
+
"eval_steps_per_second": 17.003,
|
| 284 |
+
"step": 20000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.28,
|
| 288 |
+
"learning_rate": 4.5334462165901974e-05,
|
| 289 |
+
"loss": 0.6397,
|
| 290 |
+
"step": 20500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.29,
|
| 294 |
+
"learning_rate": 4.522066856019227e-05,
|
| 295 |
+
"loss": 0.65,
|
| 296 |
+
"step": 21000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.29,
|
| 300 |
+
"learning_rate": 4.510687495448256e-05,
|
| 301 |
+
"loss": 0.6432,
|
| 302 |
+
"step": 21500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.3,
|
| 306 |
+
"learning_rate": 4.499308134877285e-05,
|
| 307 |
+
"loss": 0.6377,
|
| 308 |
+
"step": 22000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.31,
|
| 312 |
+
"learning_rate": 4.487928774306314e-05,
|
| 313 |
+
"loss": 0.6321,
|
| 314 |
+
"step": 22500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.31,
|
| 318 |
+
"learning_rate": 4.4765494137353433e-05,
|
| 319 |
+
"loss": 0.63,
|
| 320 |
+
"step": 23000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.32,
|
| 324 |
+
"learning_rate": 4.4651700531643724e-05,
|
| 325 |
+
"loss": 0.6247,
|
| 326 |
+
"step": 23500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.33,
|
| 330 |
+
"learning_rate": 4.453790692593402e-05,
|
| 331 |
+
"loss": 0.6309,
|
| 332 |
+
"step": 24000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.33,
|
| 336 |
+
"learning_rate": 4.442411332022431e-05,
|
| 337 |
+
"loss": 0.6249,
|
| 338 |
+
"step": 24500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.34,
|
| 342 |
+
"learning_rate": 4.43103197145146e-05,
|
| 343 |
+
"loss": 0.6201,
|
| 344 |
+
"step": 25000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.34,
|
| 348 |
+
"eval_accuracy": 0.8697080298884816,
|
| 349 |
+
"eval_loss": 0.568393349647522,
|
| 350 |
+
"eval_runtime": 2156.1617,
|
| 351 |
+
"eval_samples_per_second": 68.02,
|
| 352 |
+
"eval_steps_per_second": 17.005,
|
| 353 |
+
"step": 25000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.35,
|
| 357 |
+
"learning_rate": 4.41965261088049e-05,
|
| 358 |
+
"loss": 0.6113,
|
| 359 |
+
"step": 25500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.36,
|
| 363 |
+
"learning_rate": 4.408273250309518e-05,
|
| 364 |
+
"loss": 0.6106,
|
| 365 |
+
"step": 26000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.36,
|
| 369 |
+
"learning_rate": 4.396893889738548e-05,
|
| 370 |
+
"loss": 0.6125,
|
| 371 |
+
"step": 26500
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.37,
|
| 375 |
+
"learning_rate": 4.385514529167577e-05,
|
| 376 |
+
"loss": 0.6086,
|
| 377 |
+
"step": 27000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.38,
|
| 381 |
+
"learning_rate": 4.374135168596606e-05,
|
| 382 |
+
"loss": 0.6058,
|
| 383 |
+
"step": 27500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.38,
|
| 387 |
+
"learning_rate": 4.362755808025636e-05,
|
| 388 |
+
"loss": 0.5997,
|
| 389 |
+
"step": 28000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.39,
|
| 393 |
+
"learning_rate": 4.351376447454665e-05,
|
| 394 |
+
"loss": 0.5993,
|
| 395 |
+
"step": 28500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4,
|
| 399 |
+
"learning_rate": 4.339997086883694e-05,
|
| 400 |
+
"loss": 0.5941,
|
| 401 |
+
"step": 29000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.4,
|
| 405 |
+
"learning_rate": 4.328617726312723e-05,
|
| 406 |
+
"loss": 0.5986,
|
| 407 |
+
"step": 29500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.41,
|
| 411 |
+
"learning_rate": 4.317238365741753e-05,
|
| 412 |
+
"loss": 0.599,
|
| 413 |
+
"step": 30000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.41,
|
| 417 |
+
"eval_accuracy": 0.8734989669528743,
|
| 418 |
+
"eval_loss": 0.5458448529243469,
|
| 419 |
+
"eval_runtime": 2154.762,
|
| 420 |
+
"eval_samples_per_second": 68.065,
|
| 421 |
+
"eval_steps_per_second": 17.016,
|
| 422 |
+
"step": 30000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.42,
|
| 426 |
+
"learning_rate": 4.305859005170781e-05,
|
| 427 |
+
"loss": 0.5963,
|
| 428 |
+
"step": 30500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.42,
|
| 432 |
+
"learning_rate": 4.294479644599811e-05,
|
| 433 |
+
"loss": 0.5888,
|
| 434 |
+
"step": 31000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.43,
|
| 438 |
+
"learning_rate": 4.28310028402884e-05,
|
| 439 |
+
"loss": 0.5874,
|
| 440 |
+
"step": 31500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.44,
|
| 444 |
+
"learning_rate": 4.271720923457869e-05,
|
| 445 |
+
"loss": 0.5874,
|
| 446 |
+
"step": 32000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.44,
|
| 450 |
+
"learning_rate": 4.260341562886899e-05,
|
| 451 |
+
"loss": 0.5826,
|
| 452 |
+
"step": 32500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.45,
|
| 456 |
+
"learning_rate": 4.248962202315928e-05,
|
| 457 |
+
"loss": 0.5841,
|
| 458 |
+
"step": 33000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.46,
|
| 462 |
+
"learning_rate": 4.237582841744957e-05,
|
| 463 |
+
"loss": 0.5787,
|
| 464 |
+
"step": 33500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.46,
|
| 468 |
+
"learning_rate": 4.226203481173986e-05,
|
| 469 |
+
"loss": 0.586,
|
| 470 |
+
"step": 34000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.47,
|
| 474 |
+
"learning_rate": 4.2148241206030156e-05,
|
| 475 |
+
"loss": 0.575,
|
| 476 |
+
"step": 34500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.48,
|
| 480 |
+
"learning_rate": 4.203444760032044e-05,
|
| 481 |
+
"loss": 0.5736,
|
| 482 |
+
"step": 35000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.48,
|
| 486 |
+
"eval_accuracy": 0.8767297871240187,
|
| 487 |
+
"eval_loss": 0.5289933681488037,
|
| 488 |
+
"eval_runtime": 2154.7351,
|
| 489 |
+
"eval_samples_per_second": 68.065,
|
| 490 |
+
"eval_steps_per_second": 17.016,
|
| 491 |
+
"step": 35000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.48,
|
| 495 |
+
"learning_rate": 4.192065399461074e-05,
|
| 496 |
+
"loss": 0.5777,
|
| 497 |
+
"step": 35500
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.49,
|
| 501 |
+
"learning_rate": 4.1806860388901034e-05,
|
| 502 |
+
"loss": 0.5731,
|
| 503 |
+
"step": 36000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.5,
|
| 507 |
+
"learning_rate": 4.169306678319132e-05,
|
| 508 |
+
"loss": 0.5643,
|
| 509 |
+
"step": 36500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.51,
|
| 513 |
+
"learning_rate": 4.1579273177481615e-05,
|
| 514 |
+
"loss": 0.5686,
|
| 515 |
+
"step": 37000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.51,
|
| 519 |
+
"learning_rate": 4.1465479571771906e-05,
|
| 520 |
+
"loss": 0.5703,
|
| 521 |
+
"step": 37500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.52,
|
| 525 |
+
"learning_rate": 4.1351685966062196e-05,
|
| 526 |
+
"loss": 0.5681,
|
| 527 |
+
"step": 38000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.53,
|
| 531 |
+
"learning_rate": 4.1237892360352487e-05,
|
| 532 |
+
"loss": 0.5633,
|
| 533 |
+
"step": 38500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.53,
|
| 537 |
+
"learning_rate": 4.1124098754642784e-05,
|
| 538 |
+
"loss": 0.5637,
|
| 539 |
+
"step": 39000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.54,
|
| 543 |
+
"learning_rate": 4.1010305148933074e-05,
|
| 544 |
+
"loss": 0.5595,
|
| 545 |
+
"step": 39500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.55,
|
| 549 |
+
"learning_rate": 4.0896511543223365e-05,
|
| 550 |
+
"loss": 0.562,
|
| 551 |
+
"step": 40000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.55,
|
| 555 |
+
"eval_accuracy": 0.8793699356208747,
|
| 556 |
+
"eval_loss": 0.5154067873954773,
|
| 557 |
+
"eval_runtime": 2155.4611,
|
| 558 |
+
"eval_samples_per_second": 68.043,
|
| 559 |
+
"eval_steps_per_second": 17.011,
|
| 560 |
+
"step": 40000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.55,
|
| 564 |
+
"learning_rate": 4.078271793751366e-05,
|
| 565 |
+
"loss": 0.5635,
|
| 566 |
+
"step": 40500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.56,
|
| 570 |
+
"learning_rate": 4.0668924331803946e-05,
|
| 571 |
+
"loss": 0.5593,
|
| 572 |
+
"step": 41000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.57,
|
| 576 |
+
"learning_rate": 4.055513072609424e-05,
|
| 577 |
+
"loss": 0.5502,
|
| 578 |
+
"step": 41500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.57,
|
| 582 |
+
"learning_rate": 4.0441337120384534e-05,
|
| 583 |
+
"loss": 0.5559,
|
| 584 |
+
"step": 42000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.58,
|
| 588 |
+
"learning_rate": 4.0327543514674824e-05,
|
| 589 |
+
"loss": 0.5548,
|
| 590 |
+
"step": 42500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.59,
|
| 594 |
+
"learning_rate": 4.0213749908965115e-05,
|
| 595 |
+
"loss": 0.5526,
|
| 596 |
+
"step": 43000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.59,
|
| 600 |
+
"learning_rate": 4.009995630325541e-05,
|
| 601 |
+
"loss": 0.5511,
|
| 602 |
+
"step": 43500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.6,
|
| 606 |
+
"learning_rate": 3.99861626975457e-05,
|
| 607 |
+
"loss": 0.5479,
|
| 608 |
+
"step": 44000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.61,
|
| 612 |
+
"learning_rate": 3.987236909183599e-05,
|
| 613 |
+
"loss": 0.5484,
|
| 614 |
+
"step": 44500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.61,
|
| 618 |
+
"learning_rate": 3.975857548612629e-05,
|
| 619 |
+
"loss": 0.5448,
|
| 620 |
+
"step": 45000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.61,
|
| 624 |
+
"eval_accuracy": 0.8815636591304047,
|
| 625 |
+
"eval_loss": 0.5042124390602112,
|
| 626 |
+
"eval_runtime": 2154.4288,
|
| 627 |
+
"eval_samples_per_second": 68.075,
|
| 628 |
+
"eval_steps_per_second": 17.019,
|
| 629 |
+
"step": 45000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.62,
|
| 633 |
+
"learning_rate": 3.9644781880416574e-05,
|
| 634 |
+
"loss": 0.5464,
|
| 635 |
+
"step": 45500
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.63,
|
| 639 |
+
"learning_rate": 3.953098827470687e-05,
|
| 640 |
+
"loss": 0.5426,
|
| 641 |
+
"step": 46000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.63,
|
| 645 |
+
"learning_rate": 3.941719466899716e-05,
|
| 646 |
+
"loss": 0.5408,
|
| 647 |
+
"step": 46500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.64,
|
| 651 |
+
"learning_rate": 3.930340106328745e-05,
|
| 652 |
+
"loss": 0.5421,
|
| 653 |
+
"step": 47000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.65,
|
| 657 |
+
"learning_rate": 3.918960745757775e-05,
|
| 658 |
+
"loss": 0.5421,
|
| 659 |
+
"step": 47500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.66,
|
| 663 |
+
"learning_rate": 3.907581385186804e-05,
|
| 664 |
+
"loss": 0.5427,
|
| 665 |
+
"step": 48000
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.66,
|
| 669 |
+
"learning_rate": 3.896202024615833e-05,
|
| 670 |
+
"loss": 0.5434,
|
| 671 |
+
"step": 48500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.67,
|
| 675 |
+
"learning_rate": 3.884822664044862e-05,
|
| 676 |
+
"loss": 0.5364,
|
| 677 |
+
"step": 49000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.68,
|
| 681 |
+
"learning_rate": 3.873443303473892e-05,
|
| 682 |
+
"loss": 0.5382,
|
| 683 |
+
"step": 49500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.68,
|
| 687 |
+
"learning_rate": 3.86206394290292e-05,
|
| 688 |
+
"loss": 0.5354,
|
| 689 |
+
"step": 50000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.68,
|
| 693 |
+
"eval_accuracy": 0.8837189220054034,
|
| 694 |
+
"eval_loss": 0.4926992356777191,
|
| 695 |
+
"eval_runtime": 2154.3764,
|
| 696 |
+
"eval_samples_per_second": 68.077,
|
| 697 |
+
"eval_steps_per_second": 17.019,
|
| 698 |
+
"step": 50000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.69,
|
| 702 |
+
"learning_rate": 3.85068458233195e-05,
|
| 703 |
+
"loss": 0.5354,
|
| 704 |
+
"step": 50500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.7,
|
| 708 |
+
"learning_rate": 3.839305221760979e-05,
|
| 709 |
+
"loss": 0.5347,
|
| 710 |
+
"step": 51000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7,
|
| 714 |
+
"learning_rate": 3.827925861190008e-05,
|
| 715 |
+
"loss": 0.5345,
|
| 716 |
+
"step": 51500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.71,
|
| 720 |
+
"learning_rate": 3.816546500619038e-05,
|
| 721 |
+
"loss": 0.5321,
|
| 722 |
+
"step": 52000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.72,
|
| 726 |
+
"learning_rate": 3.805167140048067e-05,
|
| 727 |
+
"loss": 0.5346,
|
| 728 |
+
"step": 52500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.72,
|
| 732 |
+
"learning_rate": 3.793787779477096e-05,
|
| 733 |
+
"loss": 0.5316,
|
| 734 |
+
"step": 53000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.73,
|
| 738 |
+
"learning_rate": 3.782408418906125e-05,
|
| 739 |
+
"loss": 0.5249,
|
| 740 |
+
"step": 53500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.74,
|
| 744 |
+
"learning_rate": 3.771029058335154e-05,
|
| 745 |
+
"loss": 0.5293,
|
| 746 |
+
"step": 54000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.74,
|
| 750 |
+
"learning_rate": 3.759649697764183e-05,
|
| 751 |
+
"loss": 0.5262,
|
| 752 |
+
"step": 54500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.75,
|
| 756 |
+
"learning_rate": 3.748270337193213e-05,
|
| 757 |
+
"loss": 0.5231,
|
| 758 |
+
"step": 55000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.75,
|
| 762 |
+
"eval_accuracy": 0.88523342323253,
|
| 763 |
+
"eval_loss": 0.48392435908317566,
|
| 764 |
+
"eval_runtime": 2153.8585,
|
| 765 |
+
"eval_samples_per_second": 68.093,
|
| 766 |
+
"eval_steps_per_second": 17.023,
|
| 767 |
+
"step": 55000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.76,
|
| 771 |
+
"learning_rate": 3.736890976622242e-05,
|
| 772 |
+
"loss": 0.5265,
|
| 773 |
+
"step": 55500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.76,
|
| 777 |
+
"learning_rate": 3.725511616051271e-05,
|
| 778 |
+
"loss": 0.5316,
|
| 779 |
+
"step": 56000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.77,
|
| 783 |
+
"learning_rate": 3.7141322554803006e-05,
|
| 784 |
+
"loss": 0.5224,
|
| 785 |
+
"step": 56500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.78,
|
| 789 |
+
"learning_rate": 3.7027528949093296e-05,
|
| 790 |
+
"loss": 0.5221,
|
| 791 |
+
"step": 57000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.79,
|
| 795 |
+
"learning_rate": 3.691373534338359e-05,
|
| 796 |
+
"loss": 0.5216,
|
| 797 |
+
"step": 57500
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.79,
|
| 801 |
+
"learning_rate": 3.679994173767388e-05,
|
| 802 |
+
"loss": 0.5193,
|
| 803 |
+
"step": 58000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.8,
|
| 807 |
+
"learning_rate": 3.668614813196417e-05,
|
| 808 |
+
"loss": 0.5201,
|
| 809 |
+
"step": 58500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.81,
|
| 813 |
+
"learning_rate": 3.657235452625446e-05,
|
| 814 |
+
"loss": 0.5174,
|
| 815 |
+
"step": 59000
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.81,
|
| 819 |
+
"learning_rate": 3.6458560920544756e-05,
|
| 820 |
+
"loss": 0.5158,
|
| 821 |
+
"step": 59500
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.82,
|
| 825 |
+
"learning_rate": 3.6344767314835046e-05,
|
| 826 |
+
"loss": 0.5145,
|
| 827 |
+
"step": 60000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.82,
|
| 831 |
+
"eval_accuracy": 0.8870722746070282,
|
| 832 |
+
"eval_loss": 0.47552376985549927,
|
| 833 |
+
"eval_runtime": 2156.2198,
|
| 834 |
+
"eval_samples_per_second": 68.019,
|
| 835 |
+
"eval_steps_per_second": 17.005,
|
| 836 |
+
"step": 60000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.83,
|
| 840 |
+
"learning_rate": 3.6230973709125337e-05,
|
| 841 |
+
"loss": 0.5135,
|
| 842 |
+
"step": 60500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.83,
|
| 846 |
+
"learning_rate": 3.6117180103415634e-05,
|
| 847 |
+
"loss": 0.5207,
|
| 848 |
+
"step": 61000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.84,
|
| 852 |
+
"learning_rate": 3.600338649770592e-05,
|
| 853 |
+
"loss": 0.5184,
|
| 854 |
+
"step": 61500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.85,
|
| 858 |
+
"learning_rate": 3.5889592891996215e-05,
|
| 859 |
+
"loss": 0.5129,
|
| 860 |
+
"step": 62000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.85,
|
| 864 |
+
"learning_rate": 3.5775799286286505e-05,
|
| 865 |
+
"loss": 0.51,
|
| 866 |
+
"step": 62500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.86,
|
| 870 |
+
"learning_rate": 3.5662005680576796e-05,
|
| 871 |
+
"loss": 0.5112,
|
| 872 |
+
"step": 63000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.87,
|
| 876 |
+
"learning_rate": 3.554821207486709e-05,
|
| 877 |
+
"loss": 0.5109,
|
| 878 |
+
"step": 63500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.87,
|
| 882 |
+
"learning_rate": 3.5434418469157384e-05,
|
| 883 |
+
"loss": 0.5081,
|
| 884 |
+
"step": 64000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.88,
|
| 888 |
+
"learning_rate": 3.5320624863447674e-05,
|
| 889 |
+
"loss": 0.5121,
|
| 890 |
+
"step": 64500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.89,
|
| 894 |
+
"learning_rate": 3.5206831257737965e-05,
|
| 895 |
+
"loss": 0.5054,
|
| 896 |
+
"step": 65000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.89,
|
| 900 |
+
"eval_accuracy": 0.8885442859495518,
|
| 901 |
+
"eval_loss": 0.4669877886772156,
|
| 902 |
+
"eval_runtime": 2155.7499,
|
| 903 |
+
"eval_samples_per_second": 68.033,
|
| 904 |
+
"eval_steps_per_second": 17.008,
|
| 905 |
+
"step": 65000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.89,
|
| 909 |
+
"learning_rate": 3.509303765202826e-05,
|
| 910 |
+
"loss": 0.5039,
|
| 911 |
+
"step": 65500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.9,
|
| 915 |
+
"learning_rate": 3.4979244046318546e-05,
|
| 916 |
+
"loss": 0.5039,
|
| 917 |
+
"step": 66000
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 0.91,
|
| 921 |
+
"learning_rate": 3.486545044060884e-05,
|
| 922 |
+
"loss": 0.5036,
|
| 923 |
+
"step": 66500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.91,
|
| 927 |
+
"learning_rate": 3.4751656834899134e-05,
|
| 928 |
+
"loss": 0.5061,
|
| 929 |
+
"step": 67000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.92,
|
| 933 |
+
"learning_rate": 3.4637863229189424e-05,
|
| 934 |
+
"loss": 0.5016,
|
| 935 |
+
"step": 67500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.93,
|
| 939 |
+
"learning_rate": 3.452406962347972e-05,
|
| 940 |
+
"loss": 0.5072,
|
| 941 |
+
"step": 68000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.94,
|
| 945 |
+
"learning_rate": 3.441027601777001e-05,
|
| 946 |
+
"loss": 0.4996,
|
| 947 |
+
"step": 68500
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.94,
|
| 951 |
+
"learning_rate": 3.42964824120603e-05,
|
| 952 |
+
"loss": 0.5044,
|
| 953 |
+
"step": 69000
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.95,
|
| 957 |
+
"learning_rate": 3.418268880635059e-05,
|
| 958 |
+
"loss": 0.5011,
|
| 959 |
+
"step": 69500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.96,
|
| 963 |
+
"learning_rate": 3.406889520064089e-05,
|
| 964 |
+
"loss": 0.5025,
|
| 965 |
+
"step": 70000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.96,
|
| 969 |
+
"eval_accuracy": 0.8897554085084038,
|
| 970 |
+
"eval_loss": 0.46060749888420105,
|
| 971 |
+
"eval_runtime": 2156.7123,
|
| 972 |
+
"eval_samples_per_second": 68.003,
|
| 973 |
+
"eval_steps_per_second": 17.001,
|
| 974 |
+
"step": 70000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.96,
|
| 978 |
+
"learning_rate": 3.3955101594931174e-05,
|
| 979 |
+
"loss": 0.5004,
|
| 980 |
+
"step": 70500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.97,
|
| 984 |
+
"learning_rate": 3.384130798922147e-05,
|
| 985 |
+
"loss": 0.5013,
|
| 986 |
+
"step": 71000
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.98,
|
| 990 |
+
"learning_rate": 3.372751438351177e-05,
|
| 991 |
+
"loss": 0.504,
|
| 992 |
+
"step": 71500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.98,
|
| 996 |
+
"learning_rate": 3.361372077780205e-05,
|
| 997 |
+
"loss": 0.501,
|
| 998 |
+
"step": 72000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.99,
|
| 1002 |
+
"learning_rate": 3.349992717209235e-05,
|
| 1003 |
+
"loss": 0.4924,
|
| 1004 |
+
"step": 72500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.0,
|
| 1008 |
+
"learning_rate": 3.338613356638264e-05,
|
| 1009 |
+
"loss": 0.4979,
|
| 1010 |
+
"step": 73000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.0,
|
| 1014 |
+
"learning_rate": 3.327233996067293e-05,
|
| 1015 |
+
"loss": 0.493,
|
| 1016 |
+
"step": 73500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.01,
|
| 1020 |
+
"learning_rate": 3.315854635496322e-05,
|
| 1021 |
+
"loss": 0.4954,
|
| 1022 |
+
"step": 74000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.02,
|
| 1026 |
+
"learning_rate": 3.304475274925352e-05,
|
| 1027 |
+
"loss": 0.4975,
|
| 1028 |
+
"step": 74500
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.02,
|
| 1032 |
+
"learning_rate": 3.293095914354381e-05,
|
| 1033 |
+
"loss": 0.4931,
|
| 1034 |
+
"step": 75000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.02,
|
| 1038 |
+
"eval_accuracy": 0.8908734163811548,
|
| 1039 |
+
"eval_loss": 0.45606353878974915,
|
| 1040 |
+
"eval_runtime": 2158.0121,
|
| 1041 |
+
"eval_samples_per_second": 67.962,
|
| 1042 |
+
"eval_steps_per_second": 16.991,
|
| 1043 |
+
"step": 75000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.03,
|
| 1047 |
+
"learning_rate": 3.28171655378341e-05,
|
| 1048 |
+
"loss": 0.4948,
|
| 1049 |
+
"step": 75500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 1.04,
|
| 1053 |
+
"learning_rate": 3.2703371932124397e-05,
|
| 1054 |
+
"loss": 0.4896,
|
| 1055 |
+
"step": 76000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.04,
|
| 1059 |
+
"learning_rate": 3.258957832641468e-05,
|
| 1060 |
+
"loss": 0.4972,
|
| 1061 |
+
"step": 76500
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.05,
|
| 1065 |
+
"learning_rate": 3.247578472070498e-05,
|
| 1066 |
+
"loss": 0.4929,
|
| 1067 |
+
"step": 77000
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.06,
|
| 1071 |
+
"learning_rate": 3.236199111499527e-05,
|
| 1072 |
+
"loss": 0.4916,
|
| 1073 |
+
"step": 77500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.07,
|
| 1077 |
+
"learning_rate": 3.224819750928556e-05,
|
| 1078 |
+
"loss": 0.4898,
|
| 1079 |
+
"step": 78000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.07,
|
| 1083 |
+
"learning_rate": 3.213440390357585e-05,
|
| 1084 |
+
"loss": 0.4823,
|
| 1085 |
+
"step": 78500
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 1.08,
|
| 1089 |
+
"learning_rate": 3.2020610297866146e-05,
|
| 1090 |
+
"loss": 0.4862,
|
| 1091 |
+
"step": 79000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.09,
|
| 1095 |
+
"learning_rate": 3.190681669215644e-05,
|
| 1096 |
+
"loss": 0.4874,
|
| 1097 |
+
"step": 79500
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.09,
|
| 1101 |
+
"learning_rate": 3.179302308644673e-05,
|
| 1102 |
+
"loss": 0.4843,
|
| 1103 |
+
"step": 80000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.09,
|
| 1107 |
+
"eval_accuracy": 0.892281354181696,
|
| 1108 |
+
"eval_loss": 0.44839057326316833,
|
| 1109 |
+
"eval_runtime": 2158.2709,
|
| 1110 |
+
"eval_samples_per_second": 67.954,
|
| 1111 |
+
"eval_steps_per_second": 16.989,
|
| 1112 |
+
"step": 80000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.1,
|
| 1116 |
+
"learning_rate": 3.1679229480737025e-05,
|
| 1117 |
+
"loss": 0.4846,
|
| 1118 |
+
"step": 80500
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.11,
|
| 1122 |
+
"learning_rate": 3.156543587502731e-05,
|
| 1123 |
+
"loss": 0.4882,
|
| 1124 |
+
"step": 81000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.11,
|
| 1128 |
+
"learning_rate": 3.1451642269317606e-05,
|
| 1129 |
+
"loss": 0.4808,
|
| 1130 |
+
"step": 81500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.12,
|
| 1134 |
+
"learning_rate": 3.1337848663607896e-05,
|
| 1135 |
+
"loss": 0.4862,
|
| 1136 |
+
"step": 82000
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.13,
|
| 1140 |
+
"learning_rate": 3.122405505789819e-05,
|
| 1141 |
+
"loss": 0.483,
|
| 1142 |
+
"step": 82500
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.13,
|
| 1146 |
+
"learning_rate": 3.1110261452188484e-05,
|
| 1147 |
+
"loss": 0.4828,
|
| 1148 |
+
"step": 83000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 1.14,
|
| 1152 |
+
"learning_rate": 3.0996467846478774e-05,
|
| 1153 |
+
"loss": 0.4804,
|
| 1154 |
+
"step": 83500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.15,
|
| 1158 |
+
"learning_rate": 3.0882674240769065e-05,
|
| 1159 |
+
"loss": 0.4807,
|
| 1160 |
+
"step": 84000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.15,
|
| 1164 |
+
"learning_rate": 3.0768880635059355e-05,
|
| 1165 |
+
"loss": 0.4808,
|
| 1166 |
+
"step": 84500
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.16,
|
| 1170 |
+
"learning_rate": 3.065508702934965e-05,
|
| 1171 |
+
"loss": 0.4801,
|
| 1172 |
+
"step": 85000
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.16,
|
| 1176 |
+
"eval_accuracy": 0.8932934230811805,
|
| 1177 |
+
"eval_loss": 0.44512268900871277,
|
| 1178 |
+
"eval_runtime": 2157.4288,
|
| 1179 |
+
"eval_samples_per_second": 67.98,
|
| 1180 |
+
"eval_steps_per_second": 16.995,
|
| 1181 |
+
"step": 85000
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 1.17,
|
| 1185 |
+
"learning_rate": 3.0541293423639936e-05,
|
| 1186 |
+
"loss": 0.4768,
|
| 1187 |
+
"step": 85500
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.17,
|
| 1191 |
+
"learning_rate": 3.0427499817930234e-05,
|
| 1192 |
+
"loss": 0.4794,
|
| 1193 |
+
"step": 86000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.18,
|
| 1197 |
+
"learning_rate": 3.031370621222052e-05,
|
| 1198 |
+
"loss": 0.4828,
|
| 1199 |
+
"step": 86500
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.19,
|
| 1203 |
+
"learning_rate": 3.0199912606510815e-05,
|
| 1204 |
+
"loss": 0.4783,
|
| 1205 |
+
"step": 87000
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 1.19,
|
| 1209 |
+
"learning_rate": 3.0086119000801112e-05,
|
| 1210 |
+
"loss": 0.4776,
|
| 1211 |
+
"step": 87500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 1.2,
|
| 1215 |
+
"learning_rate": 2.99723253950914e-05,
|
| 1216 |
+
"loss": 0.4738,
|
| 1217 |
+
"step": 88000
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.21,
|
| 1221 |
+
"learning_rate": 2.9858531789381693e-05,
|
| 1222 |
+
"loss": 0.4741,
|
| 1223 |
+
"step": 88500
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 1.22,
|
| 1227 |
+
"learning_rate": 2.9744738183671984e-05,
|
| 1228 |
+
"loss": 0.4753,
|
| 1229 |
+
"step": 89000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.22,
|
| 1233 |
+
"learning_rate": 2.9630944577962277e-05,
|
| 1234 |
+
"loss": 0.4744,
|
| 1235 |
+
"step": 89500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.23,
|
| 1239 |
+
"learning_rate": 2.9517150972252568e-05,
|
| 1240 |
+
"loss": 0.4781,
|
| 1241 |
+
"step": 90000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.23,
|
| 1245 |
+
"eval_accuracy": 0.894076440048766,
|
| 1246 |
+
"eval_loss": 0.4388616979122162,
|
| 1247 |
+
"eval_runtime": 2159.5744,
|
| 1248 |
+
"eval_samples_per_second": 67.913,
|
| 1249 |
+
"eval_steps_per_second": 16.978,
|
| 1250 |
+
"step": 90000
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.24,
|
| 1254 |
+
"learning_rate": 2.9403357366542862e-05,
|
| 1255 |
+
"loss": 0.4716,
|
| 1256 |
+
"step": 90500
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.24,
|
| 1260 |
+
"learning_rate": 2.9289563760833156e-05,
|
| 1261 |
+
"loss": 0.4752,
|
| 1262 |
+
"step": 91000
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.25,
|
| 1266 |
+
"learning_rate": 2.9175770155123443e-05,
|
| 1267 |
+
"loss": 0.4676,
|
| 1268 |
+
"step": 91500
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 1.26,
|
| 1272 |
+
"learning_rate": 2.906197654941374e-05,
|
| 1273 |
+
"loss": 0.4706,
|
| 1274 |
+
"step": 92000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.26,
|
| 1278 |
+
"learning_rate": 2.8948182943704027e-05,
|
| 1279 |
+
"loss": 0.4695,
|
| 1280 |
+
"step": 92500
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.27,
|
| 1284 |
+
"learning_rate": 2.883438933799432e-05,
|
| 1285 |
+
"loss": 0.4735,
|
| 1286 |
+
"step": 93000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.28,
|
| 1290 |
+
"learning_rate": 2.872059573228461e-05,
|
| 1291 |
+
"loss": 0.4699,
|
| 1292 |
+
"step": 93500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.28,
|
| 1296 |
+
"learning_rate": 2.8606802126574906e-05,
|
| 1297 |
+
"loss": 0.4702,
|
| 1298 |
+
"step": 94000
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.29,
|
| 1302 |
+
"learning_rate": 2.8493008520865193e-05,
|
| 1303 |
+
"loss": 0.4719,
|
| 1304 |
+
"step": 94500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.3,
|
| 1308 |
+
"learning_rate": 2.837921491515549e-05,
|
| 1309 |
+
"loss": 0.4721,
|
| 1310 |
+
"step": 95000
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.3,
|
| 1314 |
+
"eval_accuracy": 0.8949646114588999,
|
| 1315 |
+
"eval_loss": 0.4361897110939026,
|
| 1316 |
+
"eval_runtime": 2158.3457,
|
| 1317 |
+
"eval_samples_per_second": 67.952,
|
| 1318 |
+
"eval_steps_per_second": 16.988,
|
| 1319 |
+
"step": 95000
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.3,
|
| 1323 |
+
"learning_rate": 2.8265421309445784e-05,
|
| 1324 |
+
"loss": 0.4673,
|
| 1325 |
+
"step": 95500
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.31,
|
| 1329 |
+
"learning_rate": 2.815162770373607e-05,
|
| 1330 |
+
"loss": 0.4686,
|
| 1331 |
+
"step": 96000
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 1.32,
|
| 1335 |
+
"learning_rate": 2.8037834098026365e-05,
|
| 1336 |
+
"loss": 0.468,
|
| 1337 |
+
"step": 96500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 1.32,
|
| 1341 |
+
"learning_rate": 2.7924040492316655e-05,
|
| 1342 |
+
"loss": 0.4638,
|
| 1343 |
+
"step": 97000
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.33,
|
| 1347 |
+
"learning_rate": 2.781024688660695e-05,
|
| 1348 |
+
"loss": 0.4644,
|
| 1349 |
+
"step": 97500
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.34,
|
| 1353 |
+
"learning_rate": 2.769645328089724e-05,
|
| 1354 |
+
"loss": 0.4688,
|
| 1355 |
+
"step": 98000
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.35,
|
| 1359 |
+
"learning_rate": 2.7582659675187534e-05,
|
| 1360 |
+
"loss": 0.4661,
|
| 1361 |
+
"step": 98500
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.35,
|
| 1365 |
+
"learning_rate": 2.7468866069477828e-05,
|
| 1366 |
+
"loss": 0.4649,
|
| 1367 |
+
"step": 99000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.36,
|
| 1371 |
+
"learning_rate": 2.7355072463768118e-05,
|
| 1372 |
+
"loss": 0.4723,
|
| 1373 |
+
"step": 99500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.37,
|
| 1377 |
+
"learning_rate": 2.7241278858058412e-05,
|
| 1378 |
+
"loss": 0.4676,
|
| 1379 |
+
"step": 100000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.37,
|
| 1383 |
+
"eval_accuracy": 0.8959113381567293,
|
| 1384 |
+
"eval_loss": 0.43063369393348694,
|
| 1385 |
+
"eval_runtime": 2155.3448,
|
| 1386 |
+
"eval_samples_per_second": 68.046,
|
| 1387 |
+
"eval_steps_per_second": 17.012,
|
| 1388 |
+
"step": 100000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.37,
|
| 1392 |
+
"learning_rate": 2.71274852523487e-05,
|
| 1393 |
+
"loss": 0.4612,
|
| 1394 |
+
"step": 100500
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 1.38,
|
| 1398 |
+
"learning_rate": 2.7013691646638993e-05,
|
| 1399 |
+
"loss": 0.466,
|
| 1400 |
+
"step": 101000
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.39,
|
| 1404 |
+
"learning_rate": 2.6899898040929283e-05,
|
| 1405 |
+
"loss": 0.4618,
|
| 1406 |
+
"step": 101500
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 1.39,
|
| 1410 |
+
"learning_rate": 2.6786104435219577e-05,
|
| 1411 |
+
"loss": 0.4627,
|
| 1412 |
+
"step": 102000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.4,
|
| 1416 |
+
"learning_rate": 2.6672310829509868e-05,
|
| 1417 |
+
"loss": 0.4658,
|
| 1418 |
+
"step": 102500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 1.41,
|
| 1422 |
+
"learning_rate": 2.6558517223800162e-05,
|
| 1423 |
+
"loss": 0.4598,
|
| 1424 |
+
"step": 103000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.41,
|
| 1428 |
+
"learning_rate": 2.6444723618090456e-05,
|
| 1429 |
+
"loss": 0.4641,
|
| 1430 |
+
"step": 103500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.42,
|
| 1434 |
+
"learning_rate": 2.6330930012380746e-05,
|
| 1435 |
+
"loss": 0.4544,
|
| 1436 |
+
"step": 104000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 1.43,
|
| 1440 |
+
"learning_rate": 2.621713640667104e-05,
|
| 1441 |
+
"loss": 0.4587,
|
| 1442 |
+
"step": 104500
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 1.43,
|
| 1446 |
+
"learning_rate": 2.6103342800961327e-05,
|
| 1447 |
+
"loss": 0.4569,
|
| 1448 |
+
"step": 105000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.43,
|
| 1452 |
+
"eval_accuracy": 0.8969798763719528,
|
| 1453 |
+
"eval_loss": 0.4255083203315735,
|
| 1454 |
+
"eval_runtime": 2157.7053,
|
| 1455 |
+
"eval_samples_per_second": 67.972,
|
| 1456 |
+
"eval_steps_per_second": 16.993,
|
| 1457 |
+
"step": 105000
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 1.44,
|
| 1461 |
+
"learning_rate": 2.598954919525162e-05,
|
| 1462 |
+
"loss": 0.4601,
|
| 1463 |
+
"step": 105500
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 1.45,
|
| 1467 |
+
"learning_rate": 2.587575558954191e-05,
|
| 1468 |
+
"loss": 0.4572,
|
| 1469 |
+
"step": 106000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 1.45,
|
| 1473 |
+
"learning_rate": 2.5761961983832205e-05,
|
| 1474 |
+
"loss": 0.4527,
|
| 1475 |
+
"step": 106500
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 1.46,
|
| 1479 |
+
"learning_rate": 2.56481683781225e-05,
|
| 1480 |
+
"loss": 0.4595,
|
| 1481 |
+
"step": 107000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.47,
|
| 1485 |
+
"learning_rate": 2.553437477241279e-05,
|
| 1486 |
+
"loss": 0.4621,
|
| 1487 |
+
"step": 107500
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.47,
|
| 1491 |
+
"learning_rate": 2.5420581166703084e-05,
|
| 1492 |
+
"loss": 0.458,
|
| 1493 |
+
"step": 108000
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 1.48,
|
| 1497 |
+
"learning_rate": 2.530678756099337e-05,
|
| 1498 |
+
"loss": 0.4606,
|
| 1499 |
+
"step": 108500
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 1.49,
|
| 1503 |
+
"learning_rate": 2.5192993955283668e-05,
|
| 1504 |
+
"loss": 0.4573,
|
| 1505 |
+
"step": 109000
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.5,
|
| 1509 |
+
"learning_rate": 2.5079200349573955e-05,
|
| 1510 |
+
"loss": 0.4601,
|
| 1511 |
+
"step": 109500
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 1.5,
|
| 1515 |
+
"learning_rate": 2.496540674386425e-05,
|
| 1516 |
+
"loss": 0.4562,
|
| 1517 |
+
"step": 110000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 1.5,
|
| 1521 |
+
"eval_accuracy": 0.897719716014404,
|
| 1522 |
+
"eval_loss": 0.4225170314311981,
|
| 1523 |
+
"eval_runtime": 2157.2636,
|
| 1524 |
+
"eval_samples_per_second": 67.986,
|
| 1525 |
+
"eval_steps_per_second": 16.997,
|
| 1526 |
+
"step": 110000
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 1.51,
|
| 1530 |
+
"learning_rate": 2.4851613138154543e-05,
|
| 1531 |
+
"loss": 0.4552,
|
| 1532 |
+
"step": 110500
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 1.52,
|
| 1536 |
+
"learning_rate": 2.4737819532444834e-05,
|
| 1537 |
+
"loss": 0.4577,
|
| 1538 |
+
"step": 111000
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 1.52,
|
| 1542 |
+
"learning_rate": 2.4624025926735124e-05,
|
| 1543 |
+
"loss": 0.4526,
|
| 1544 |
+
"step": 111500
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.53,
|
| 1548 |
+
"learning_rate": 2.4510232321025418e-05,
|
| 1549 |
+
"loss": 0.4533,
|
| 1550 |
+
"step": 112000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.54,
|
| 1554 |
+
"learning_rate": 2.4396438715315712e-05,
|
| 1555 |
+
"loss": 0.456,
|
| 1556 |
+
"step": 112500
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 1.54,
|
| 1560 |
+
"learning_rate": 2.4282645109606002e-05,
|
| 1561 |
+
"loss": 0.4536,
|
| 1562 |
+
"step": 113000
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 1.55,
|
| 1566 |
+
"learning_rate": 2.4168851503896296e-05,
|
| 1567 |
+
"loss": 0.4612,
|
| 1568 |
+
"step": 113500
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 1.56,
|
| 1572 |
+
"learning_rate": 2.4055057898186587e-05,
|
| 1573 |
+
"loss": 0.4527,
|
| 1574 |
+
"step": 114000
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 1.56,
|
| 1578 |
+
"learning_rate": 2.3941264292476877e-05,
|
| 1579 |
+
"loss": 0.4528,
|
| 1580 |
+
"step": 114500
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 1.57,
|
| 1584 |
+
"learning_rate": 2.382747068676717e-05,
|
| 1585 |
+
"loss": 0.4532,
|
| 1586 |
+
"step": 115000
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 1.57,
|
| 1590 |
+
"eval_accuracy": 0.8983772130502822,
|
| 1591 |
+
"eval_loss": 0.4186096489429474,
|
| 1592 |
+
"eval_runtime": 2156.0493,
|
| 1593 |
+
"eval_samples_per_second": 68.024,
|
| 1594 |
+
"eval_steps_per_second": 17.006,
|
| 1595 |
+
"step": 115000
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 1.58,
|
| 1599 |
+
"learning_rate": 2.3713677081057462e-05,
|
| 1600 |
+
"loss": 0.4508,
|
| 1601 |
+
"step": 115500
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 1.58,
|
| 1605 |
+
"learning_rate": 2.3599883475347752e-05,
|
| 1606 |
+
"loss": 0.4521,
|
| 1607 |
+
"step": 116000
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 1.59,
|
| 1611 |
+
"learning_rate": 2.3486089869638046e-05,
|
| 1612 |
+
"loss": 0.4492,
|
| 1613 |
+
"step": 116500
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.6,
|
| 1617 |
+
"learning_rate": 2.337229626392834e-05,
|
| 1618 |
+
"loss": 0.4485,
|
| 1619 |
+
"step": 117000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 1.6,
|
| 1623 |
+
"learning_rate": 2.325850265821863e-05,
|
| 1624 |
+
"loss": 0.4499,
|
| 1625 |
+
"step": 117500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 1.61,
|
| 1629 |
+
"learning_rate": 2.3144709052508924e-05,
|
| 1630 |
+
"loss": 0.4495,
|
| 1631 |
+
"step": 118000
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 1.62,
|
| 1635 |
+
"learning_rate": 2.3030915446799215e-05,
|
| 1636 |
+
"loss": 0.4499,
|
| 1637 |
+
"step": 118500
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 1.62,
|
| 1641 |
+
"learning_rate": 2.2917121841089505e-05,
|
| 1642 |
+
"loss": 0.4508,
|
| 1643 |
+
"step": 119000
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 1.63,
|
| 1647 |
+
"learning_rate": 2.28033282353798e-05,
|
| 1648 |
+
"loss": 0.4474,
|
| 1649 |
+
"step": 119500
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.64,
|
| 1653 |
+
"learning_rate": 2.268953462967009e-05,
|
| 1654 |
+
"loss": 0.4481,
|
| 1655 |
+
"step": 120000
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.64,
|
| 1659 |
+
"eval_accuracy": 0.8989962769112119,
|
| 1660 |
+
"eval_loss": 0.4165661036968231,
|
| 1661 |
+
"eval_runtime": 2156.6641,
|
| 1662 |
+
"eval_samples_per_second": 68.005,
|
| 1663 |
+
"eval_steps_per_second": 17.001,
|
| 1664 |
+
"step": 120000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 1.65,
|
| 1668 |
+
"learning_rate": 2.2575741023960384e-05,
|
| 1669 |
+
"loss": 0.4489,
|
| 1670 |
+
"step": 120500
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 1.65,
|
| 1674 |
+
"learning_rate": 2.2461947418250674e-05,
|
| 1675 |
+
"loss": 0.4509,
|
| 1676 |
+
"step": 121000
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.66,
|
| 1680 |
+
"learning_rate": 2.2348153812540968e-05,
|
| 1681 |
+
"loss": 0.4495,
|
| 1682 |
+
"step": 121500
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 1.67,
|
| 1686 |
+
"learning_rate": 2.223436020683126e-05,
|
| 1687 |
+
"loss": 0.4475,
|
| 1688 |
+
"step": 122000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.67,
|
| 1692 |
+
"learning_rate": 2.212056660112155e-05,
|
| 1693 |
+
"loss": 0.4502,
|
| 1694 |
+
"step": 122500
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.68,
|
| 1698 |
+
"learning_rate": 2.2006772995411843e-05,
|
| 1699 |
+
"loss": 0.4448,
|
| 1700 |
+
"step": 123000
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 1.69,
|
| 1704 |
+
"learning_rate": 2.1892979389702134e-05,
|
| 1705 |
+
"loss": 0.4511,
|
| 1706 |
+
"step": 123500
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 1.69,
|
| 1710 |
+
"learning_rate": 2.1779185783992424e-05,
|
| 1711 |
+
"loss": 0.4436,
|
| 1712 |
+
"step": 124000
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 1.7,
|
| 1716 |
+
"learning_rate": 2.166539217828272e-05,
|
| 1717 |
+
"loss": 0.4401,
|
| 1718 |
+
"step": 124500
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.71,
|
| 1722 |
+
"learning_rate": 2.1551598572573012e-05,
|
| 1723 |
+
"loss": 0.4474,
|
| 1724 |
+
"step": 125000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 1.71,
|
| 1728 |
+
"eval_accuracy": 0.8998707125729046,
|
| 1729 |
+
"eval_loss": 0.41221463680267334,
|
| 1730 |
+
"eval_runtime": 2156.9802,
|
| 1731 |
+
"eval_samples_per_second": 67.995,
|
| 1732 |
+
"eval_steps_per_second": 16.999,
|
| 1733 |
+
"step": 125000
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 1.71,
|
| 1737 |
+
"learning_rate": 2.1437804966863302e-05,
|
| 1738 |
+
"loss": 0.4452,
|
| 1739 |
+
"step": 125500
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.72,
|
| 1743 |
+
"learning_rate": 2.1324011361153596e-05,
|
| 1744 |
+
"loss": 0.4375,
|
| 1745 |
+
"step": 126000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 1.73,
|
| 1749 |
+
"learning_rate": 2.1210217755443887e-05,
|
| 1750 |
+
"loss": 0.4429,
|
| 1751 |
+
"step": 126500
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 1.73,
|
| 1755 |
+
"learning_rate": 2.1096424149734177e-05,
|
| 1756 |
+
"loss": 0.4426,
|
| 1757 |
+
"step": 127000
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 1.74,
|
| 1761 |
+
"learning_rate": 2.098263054402447e-05,
|
| 1762 |
+
"loss": 0.4427,
|
| 1763 |
+
"step": 127500
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 1.75,
|
| 1767 |
+
"learning_rate": 2.086883693831476e-05,
|
| 1768 |
+
"loss": 0.4434,
|
| 1769 |
+
"step": 128000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 1.75,
|
| 1773 |
+
"learning_rate": 2.0755043332605056e-05,
|
| 1774 |
+
"loss": 0.442,
|
| 1775 |
+
"step": 128500
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 1.76,
|
| 1779 |
+
"learning_rate": 2.064124972689535e-05,
|
| 1780 |
+
"loss": 0.4409,
|
| 1781 |
+
"step": 129000
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.77,
|
| 1785 |
+
"learning_rate": 2.052745612118564e-05,
|
| 1786 |
+
"loss": 0.4422,
|
| 1787 |
+
"step": 129500
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 1.78,
|
| 1791 |
+
"learning_rate": 2.041366251547593e-05,
|
| 1792 |
+
"loss": 0.4417,
|
| 1793 |
+
"step": 130000
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.78,
|
| 1797 |
+
"eval_accuracy": 0.9003224313444536,
|
| 1798 |
+
"eval_loss": 0.40945446491241455,
|
| 1799 |
+
"eval_runtime": 2157.4017,
|
| 1800 |
+
"eval_samples_per_second": 67.981,
|
| 1801 |
+
"eval_steps_per_second": 16.995,
|
| 1802 |
+
"step": 130000
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.78,
|
| 1806 |
+
"learning_rate": 2.0299868909766224e-05,
|
| 1807 |
+
"loss": 0.4395,
|
| 1808 |
+
"step": 130500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 1.79,
|
| 1812 |
+
"learning_rate": 2.0186075304056515e-05,
|
| 1813 |
+
"loss": 0.4428,
|
| 1814 |
+
"step": 131000
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 1.8,
|
| 1818 |
+
"learning_rate": 2.0072281698346805e-05,
|
| 1819 |
+
"loss": 0.4437,
|
| 1820 |
+
"step": 131500
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 1.8,
|
| 1824 |
+
"learning_rate": 1.99584880926371e-05,
|
| 1825 |
+
"loss": 0.438,
|
| 1826 |
+
"step": 132000
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 1.81,
|
| 1830 |
+
"learning_rate": 1.9844694486927393e-05,
|
| 1831 |
+
"loss": 0.4379,
|
| 1832 |
+
"step": 132500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 1.82,
|
| 1836 |
+
"learning_rate": 1.9730900881217684e-05,
|
| 1837 |
+
"loss": 0.4381,
|
| 1838 |
+
"step": 133000
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 1.82,
|
| 1842 |
+
"learning_rate": 1.9617107275507978e-05,
|
| 1843 |
+
"loss": 0.4442,
|
| 1844 |
+
"step": 133500
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.83,
|
| 1848 |
+
"learning_rate": 1.9503313669798268e-05,
|
| 1849 |
+
"loss": 0.4404,
|
| 1850 |
+
"step": 134000
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 1.84,
|
| 1854 |
+
"learning_rate": 1.938952006408856e-05,
|
| 1855 |
+
"loss": 0.4386,
|
| 1856 |
+
"step": 134500
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 1.84,
|
| 1860 |
+
"learning_rate": 1.9275726458378852e-05,
|
| 1861 |
+
"loss": 0.4329,
|
| 1862 |
+
"step": 135000
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 1.84,
|
| 1866 |
+
"eval_accuracy": 0.9011228326204502,
|
| 1867 |
+
"eval_loss": 0.4052372872829437,
|
| 1868 |
+
"eval_runtime": 2157.9151,
|
| 1869 |
+
"eval_samples_per_second": 67.965,
|
| 1870 |
+
"eval_steps_per_second": 16.991,
|
| 1871 |
+
"step": 135000
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 1.85,
|
| 1875 |
+
"learning_rate": 1.9161932852669143e-05,
|
| 1876 |
+
"loss": 0.4336,
|
| 1877 |
+
"step": 135500
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 1.86,
|
| 1881 |
+
"learning_rate": 1.9048139246959433e-05,
|
| 1882 |
+
"loss": 0.4388,
|
| 1883 |
+
"step": 136000
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.86,
|
| 1887 |
+
"learning_rate": 1.8934345641249727e-05,
|
| 1888 |
+
"loss": 0.4335,
|
| 1889 |
+
"step": 136500
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 1.87,
|
| 1893 |
+
"learning_rate": 1.882055203554002e-05,
|
| 1894 |
+
"loss": 0.4346,
|
| 1895 |
+
"step": 137000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 1.88,
|
| 1899 |
+
"learning_rate": 1.8706758429830312e-05,
|
| 1900 |
+
"loss": 0.4335,
|
| 1901 |
+
"step": 137500
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 1.88,
|
| 1905 |
+
"learning_rate": 1.8592964824120602e-05,
|
| 1906 |
+
"loss": 0.4351,
|
| 1907 |
+
"step": 138000
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.89,
|
| 1911 |
+
"learning_rate": 1.8479171218410896e-05,
|
| 1912 |
+
"loss": 0.4318,
|
| 1913 |
+
"step": 138500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 1.9,
|
| 1917 |
+
"learning_rate": 1.8365377612701187e-05,
|
| 1918 |
+
"loss": 0.4337,
|
| 1919 |
+
"step": 139000
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 1.9,
|
| 1923 |
+
"learning_rate": 1.825158400699148e-05,
|
| 1924 |
+
"loss": 0.4377,
|
| 1925 |
+
"step": 139500
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 1.91,
|
| 1929 |
+
"learning_rate": 1.8137790401281774e-05,
|
| 1930 |
+
"loss": 0.436,
|
| 1931 |
+
"step": 140000
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 1.91,
|
| 1935 |
+
"eval_accuracy": 0.9015046107759694,
|
| 1936 |
+
"eval_loss": 0.40359699726104736,
|
| 1937 |
+
"eval_runtime": 2159.0025,
|
| 1938 |
+
"eval_samples_per_second": 67.931,
|
| 1939 |
+
"eval_steps_per_second": 16.983,
|
| 1940 |
+
"step": 140000
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 1.92,
|
| 1944 |
+
"learning_rate": 1.8023996795572065e-05,
|
| 1945 |
+
"loss": 0.4322,
|
| 1946 |
+
"step": 140500
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 1.93,
|
| 1950 |
+
"learning_rate": 1.7910203189862355e-05,
|
| 1951 |
+
"loss": 0.4367,
|
| 1952 |
+
"step": 141000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 1.93,
|
| 1956 |
+
"learning_rate": 1.779640958415265e-05,
|
| 1957 |
+
"loss": 0.4289,
|
| 1958 |
+
"step": 141500
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 1.94,
|
| 1962 |
+
"learning_rate": 1.768261597844294e-05,
|
| 1963 |
+
"loss": 0.4357,
|
| 1964 |
+
"step": 142000
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 1.95,
|
| 1968 |
+
"learning_rate": 1.756882237273323e-05,
|
| 1969 |
+
"loss": 0.4316,
|
| 1970 |
+
"step": 142500
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.95,
|
| 1974 |
+
"learning_rate": 1.7455028767023524e-05,
|
| 1975 |
+
"loss": 0.4334,
|
| 1976 |
+
"step": 143000
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 1.96,
|
| 1980 |
+
"learning_rate": 1.7341235161313815e-05,
|
| 1981 |
+
"loss": 0.4298,
|
| 1982 |
+
"step": 143500
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 1.97,
|
| 1986 |
+
"learning_rate": 1.722744155560411e-05,
|
| 1987 |
+
"loss": 0.4344,
|
| 1988 |
+
"step": 144000
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 1.97,
|
| 1992 |
+
"learning_rate": 1.7113647949894403e-05,
|
| 1993 |
+
"loss": 0.4309,
|
| 1994 |
+
"step": 144500
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 1.98,
|
| 1998 |
+
"learning_rate": 1.6999854344184693e-05,
|
| 1999 |
+
"loss": 0.4289,
|
| 2000 |
+
"step": 145000
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 1.98,
|
| 2004 |
+
"eval_accuracy": 0.9022181156532487,
|
| 2005 |
+
"eval_loss": 0.4002615809440613,
|
| 2006 |
+
"eval_runtime": 2157.6915,
|
| 2007 |
+
"eval_samples_per_second": 67.972,
|
| 2008 |
+
"eval_steps_per_second": 16.993,
|
| 2009 |
+
"step": 145000
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 1.99,
|
| 2013 |
+
"learning_rate": 1.6886060738474984e-05,
|
| 2014 |
+
"loss": 0.4271,
|
| 2015 |
+
"step": 145500
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 1.99,
|
| 2019 |
+
"learning_rate": 1.6772267132765277e-05,
|
| 2020 |
+
"loss": 0.4264,
|
| 2021 |
+
"step": 146000
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 2.0,
|
| 2025 |
+
"learning_rate": 1.6658473527055568e-05,
|
| 2026 |
+
"loss": 0.43,
|
| 2027 |
+
"step": 146500
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.01,
|
| 2031 |
+
"learning_rate": 1.654467992134586e-05,
|
| 2032 |
+
"loss": 0.4292,
|
| 2033 |
+
"step": 147000
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.01,
|
| 2037 |
+
"learning_rate": 1.6430886315636152e-05,
|
| 2038 |
+
"loss": 0.429,
|
| 2039 |
+
"step": 147500
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.02,
|
| 2043 |
+
"learning_rate": 1.6317092709926446e-05,
|
| 2044 |
+
"loss": 0.4265,
|
| 2045 |
+
"step": 148000
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 2.03,
|
| 2049 |
+
"learning_rate": 1.6203299104216737e-05,
|
| 2050 |
+
"loss": 0.4283,
|
| 2051 |
+
"step": 148500
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 2.03,
|
| 2055 |
+
"learning_rate": 1.608950549850703e-05,
|
| 2056 |
+
"loss": 0.4279,
|
| 2057 |
+
"step": 149000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 2.04,
|
| 2061 |
+
"learning_rate": 1.597571189279732e-05,
|
| 2062 |
+
"loss": 0.4272,
|
| 2063 |
+
"step": 149500
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 2.05,
|
| 2067 |
+
"learning_rate": 1.586191828708761e-05,
|
| 2068 |
+
"loss": 0.4326,
|
| 2069 |
+
"step": 150000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.05,
|
| 2073 |
+
"eval_accuracy": 0.902766321286562,
|
| 2074 |
+
"eval_loss": 0.3976551294326782,
|
| 2075 |
+
"eval_runtime": 2157.9395,
|
| 2076 |
+
"eval_samples_per_second": 67.964,
|
| 2077 |
+
"eval_steps_per_second": 16.991,
|
| 2078 |
+
"step": 150000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 2.06,
|
| 2082 |
+
"learning_rate": 1.5748124681377906e-05,
|
| 2083 |
+
"loss": 0.4257,
|
| 2084 |
+
"step": 150500
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 2.06,
|
| 2088 |
+
"learning_rate": 1.5634331075668196e-05,
|
| 2089 |
+
"loss": 0.4223,
|
| 2090 |
+
"step": 151000
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.07,
|
| 2094 |
+
"learning_rate": 1.5520537469958487e-05,
|
| 2095 |
+
"loss": 0.4228,
|
| 2096 |
+
"step": 151500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.08,
|
| 2100 |
+
"learning_rate": 1.540674386424878e-05,
|
| 2101 |
+
"loss": 0.4276,
|
| 2102 |
+
"step": 152000
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.08,
|
| 2106 |
+
"learning_rate": 1.5292950258539074e-05,
|
| 2107 |
+
"loss": 0.4244,
|
| 2108 |
+
"step": 152500
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 2.09,
|
| 2112 |
+
"learning_rate": 1.5179156652829365e-05,
|
| 2113 |
+
"loss": 0.4261,
|
| 2114 |
+
"step": 153000
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 2.1,
|
| 2118 |
+
"learning_rate": 1.5065363047119657e-05,
|
| 2119 |
+
"loss": 0.4226,
|
| 2120 |
+
"step": 153500
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 2.1,
|
| 2124 |
+
"learning_rate": 1.495156944140995e-05,
|
| 2125 |
+
"loss": 0.4265,
|
| 2126 |
+
"step": 154000
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 2.11,
|
| 2130 |
+
"learning_rate": 1.483777583570024e-05,
|
| 2131 |
+
"loss": 0.4262,
|
| 2132 |
+
"step": 154500
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 2.12,
|
| 2136 |
+
"learning_rate": 1.4723982229990532e-05,
|
| 2137 |
+
"loss": 0.4231,
|
| 2138 |
+
"step": 155000
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.12,
|
| 2142 |
+
"eval_accuracy": 0.9031951195206896,
|
| 2143 |
+
"eval_loss": 0.3947906792163849,
|
| 2144 |
+
"eval_runtime": 2160.1666,
|
| 2145 |
+
"eval_samples_per_second": 67.894,
|
| 2146 |
+
"eval_steps_per_second": 16.974,
|
| 2147 |
+
"step": 155000
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 2.12,
|
| 2151 |
+
"learning_rate": 1.4610188624280824e-05,
|
| 2152 |
+
"loss": 0.4195,
|
| 2153 |
+
"step": 155500
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 2.13,
|
| 2157 |
+
"learning_rate": 1.4496395018571118e-05,
|
| 2158 |
+
"loss": 0.4216,
|
| 2159 |
+
"step": 156000
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.14,
|
| 2163 |
+
"learning_rate": 1.438260141286141e-05,
|
| 2164 |
+
"loss": 0.4152,
|
| 2165 |
+
"step": 156500
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.14,
|
| 2169 |
+
"learning_rate": 1.4268807807151702e-05,
|
| 2170 |
+
"loss": 0.4194,
|
| 2171 |
+
"step": 157000
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 2.15,
|
| 2175 |
+
"learning_rate": 1.4155014201441993e-05,
|
| 2176 |
+
"loss": 0.4193,
|
| 2177 |
+
"step": 157500
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 2.16,
|
| 2181 |
+
"learning_rate": 1.4041220595732285e-05,
|
| 2182 |
+
"loss": 0.4223,
|
| 2183 |
+
"step": 158000
|
| 2184 |
+
},
|
| 2185 |
+
{
|
| 2186 |
+
"epoch": 2.16,
|
| 2187 |
+
"learning_rate": 1.3927426990022577e-05,
|
| 2188 |
+
"loss": 0.4259,
|
| 2189 |
+
"step": 158500
|
| 2190 |
+
},
|
| 2191 |
+
{
|
| 2192 |
+
"epoch": 2.17,
|
| 2193 |
+
"learning_rate": 1.3813633384312868e-05,
|
| 2194 |
+
"loss": 0.4215,
|
| 2195 |
+
"step": 159000
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 2.18,
|
| 2199 |
+
"learning_rate": 1.369983977860316e-05,
|
| 2200 |
+
"loss": 0.4195,
|
| 2201 |
+
"step": 159500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.18,
|
| 2205 |
+
"learning_rate": 1.3586046172893454e-05,
|
| 2206 |
+
"loss": 0.4233,
|
| 2207 |
+
"step": 160000
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 2.18,
|
| 2211 |
+
"eval_accuracy": 0.9036811615656513,
|
| 2212 |
+
"eval_loss": 0.3929060399532318,
|
| 2213 |
+
"eval_runtime": 2157.8805,
|
| 2214 |
+
"eval_samples_per_second": 67.966,
|
| 2215 |
+
"eval_steps_per_second": 16.992,
|
| 2216 |
+
"step": 160000
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 2.19,
|
| 2220 |
+
"learning_rate": 1.3472252567183746e-05,
|
| 2221 |
+
"loss": 0.4216,
|
| 2222 |
+
"step": 160500
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.2,
|
| 2226 |
+
"learning_rate": 1.3358458961474038e-05,
|
| 2227 |
+
"loss": 0.4192,
|
| 2228 |
+
"step": 161000
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 2.21,
|
| 2232 |
+
"learning_rate": 1.3244665355764329e-05,
|
| 2233 |
+
"loss": 0.4202,
|
| 2234 |
+
"step": 161500
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 2.21,
|
| 2238 |
+
"learning_rate": 1.3130871750054621e-05,
|
| 2239 |
+
"loss": 0.4239,
|
| 2240 |
+
"step": 162000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 2.22,
|
| 2244 |
+
"learning_rate": 1.3017078144344913e-05,
|
| 2245 |
+
"loss": 0.4254,
|
| 2246 |
+
"step": 162500
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"epoch": 2.23,
|
| 2250 |
+
"learning_rate": 1.2903284538635205e-05,
|
| 2251 |
+
"loss": 0.4154,
|
| 2252 |
+
"step": 163000
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 2.23,
|
| 2256 |
+
"learning_rate": 1.2789490932925496e-05,
|
| 2257 |
+
"loss": 0.4232,
|
| 2258 |
+
"step": 163500
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 2.24,
|
| 2262 |
+
"learning_rate": 1.2675697327215792e-05,
|
| 2263 |
+
"loss": 0.422,
|
| 2264 |
+
"step": 164000
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.25,
|
| 2268 |
+
"learning_rate": 1.2561903721506082e-05,
|
| 2269 |
+
"loss": 0.4177,
|
| 2270 |
+
"step": 164500
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.25,
|
| 2274 |
+
"learning_rate": 1.2448110115796374e-05,
|
| 2275 |
+
"loss": 0.4179,
|
| 2276 |
+
"step": 165000
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 2.25,
|
| 2280 |
+
"eval_accuracy": 0.9042119348171125,
|
| 2281 |
+
"eval_loss": 0.3903782367706299,
|
| 2282 |
+
"eval_runtime": 2158.6911,
|
| 2283 |
+
"eval_samples_per_second": 67.941,
|
| 2284 |
+
"eval_steps_per_second": 16.985,
|
| 2285 |
+
"step": 165000
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.26,
|
| 2289 |
+
"learning_rate": 1.2334316510086666e-05,
|
| 2290 |
+
"loss": 0.4215,
|
| 2291 |
+
"step": 165500
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 2.27,
|
| 2295 |
+
"learning_rate": 1.2220522904376957e-05,
|
| 2296 |
+
"loss": 0.4206,
|
| 2297 |
+
"step": 166000
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 2.27,
|
| 2301 |
+
"learning_rate": 1.210672929866725e-05,
|
| 2302 |
+
"loss": 0.4186,
|
| 2303 |
+
"step": 166500
|
| 2304 |
+
},
|
| 2305 |
+
{
|
| 2306 |
+
"epoch": 2.28,
|
| 2307 |
+
"learning_rate": 1.1992935692957543e-05,
|
| 2308 |
+
"loss": 0.4205,
|
| 2309 |
+
"step": 167000
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 2.29,
|
| 2313 |
+
"learning_rate": 1.1879142087247834e-05,
|
| 2314 |
+
"loss": 0.4163,
|
| 2315 |
+
"step": 167500
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 2.29,
|
| 2319 |
+
"learning_rate": 1.1765348481538126e-05,
|
| 2320 |
+
"loss": 0.4178,
|
| 2321 |
+
"step": 168000
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.3,
|
| 2325 |
+
"learning_rate": 1.1651554875828418e-05,
|
| 2326 |
+
"loss": 0.4163,
|
| 2327 |
+
"step": 168500
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.31,
|
| 2331 |
+
"learning_rate": 1.153776127011871e-05,
|
| 2332 |
+
"loss": 0.4176,
|
| 2333 |
+
"step": 169000
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 2.31,
|
| 2337 |
+
"learning_rate": 1.1423967664409002e-05,
|
| 2338 |
+
"loss": 0.4179,
|
| 2339 |
+
"step": 169500
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 2.32,
|
| 2343 |
+
"learning_rate": 1.1310174058699295e-05,
|
| 2344 |
+
"loss": 0.4141,
|
| 2345 |
+
"step": 170000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 2.32,
|
| 2349 |
+
"eval_accuracy": 0.9047012480604113,
|
| 2350 |
+
"eval_loss": 0.38813966512680054,
|
| 2351 |
+
"eval_runtime": 2158.6653,
|
| 2352 |
+
"eval_samples_per_second": 67.942,
|
| 2353 |
+
"eval_steps_per_second": 16.985,
|
| 2354 |
+
"step": 170000
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 2.33,
|
| 2358 |
+
"learning_rate": 1.1196380452989585e-05,
|
| 2359 |
+
"loss": 0.4174,
|
| 2360 |
+
"step": 170500
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 2.34,
|
| 2364 |
+
"learning_rate": 1.1082586847279879e-05,
|
| 2365 |
+
"loss": 0.4172,
|
| 2366 |
+
"step": 171000
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 2.34,
|
| 2370 |
+
"learning_rate": 1.096879324157017e-05,
|
| 2371 |
+
"loss": 0.4194,
|
| 2372 |
+
"step": 171500
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 2.35,
|
| 2376 |
+
"learning_rate": 1.0854999635860462e-05,
|
| 2377 |
+
"loss": 0.4147,
|
| 2378 |
+
"step": 172000
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 2.36,
|
| 2382 |
+
"learning_rate": 1.0741206030150754e-05,
|
| 2383 |
+
"loss": 0.4104,
|
| 2384 |
+
"step": 172500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 2.36,
|
| 2388 |
+
"learning_rate": 1.0627412424441046e-05,
|
| 2389 |
+
"loss": 0.4128,
|
| 2390 |
+
"step": 173000
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.37,
|
| 2394 |
+
"learning_rate": 1.0513618818731338e-05,
|
| 2395 |
+
"loss": 0.4111,
|
| 2396 |
+
"step": 173500
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 2.38,
|
| 2400 |
+
"learning_rate": 1.039982521302163e-05,
|
| 2401 |
+
"loss": 0.4159,
|
| 2402 |
+
"step": 174000
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 2.38,
|
| 2406 |
+
"learning_rate": 1.0286031607311923e-05,
|
| 2407 |
+
"loss": 0.4163,
|
| 2408 |
+
"step": 174500
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 2.39,
|
| 2412 |
+
"learning_rate": 1.0172238001602215e-05,
|
| 2413 |
+
"loss": 0.4138,
|
| 2414 |
+
"step": 175000
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 2.39,
|
| 2418 |
+
"eval_accuracy": 0.9050543084721216,
|
| 2419 |
+
"eval_loss": 0.3864535987377167,
|
| 2420 |
+
"eval_runtime": 2159.9244,
|
| 2421 |
+
"eval_samples_per_second": 67.902,
|
| 2422 |
+
"eval_steps_per_second": 16.976,
|
| 2423 |
+
"step": 175000
|
| 2424 |
+
},
|
| 2425 |
+
{
|
| 2426 |
+
"epoch": 2.4,
|
| 2427 |
+
"learning_rate": 1.0058444395892507e-05,
|
| 2428 |
+
"loss": 0.4141,
|
| 2429 |
+
"step": 175500
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 2.4,
|
| 2433 |
+
"learning_rate": 9.944650790182798e-06,
|
| 2434 |
+
"loss": 0.4129,
|
| 2435 |
+
"step": 176000
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 2.41,
|
| 2439 |
+
"learning_rate": 9.830857184473091e-06,
|
| 2440 |
+
"loss": 0.4119,
|
| 2441 |
+
"step": 176500
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 2.42,
|
| 2445 |
+
"learning_rate": 9.717063578763384e-06,
|
| 2446 |
+
"loss": 0.4067,
|
| 2447 |
+
"step": 177000
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.42,
|
| 2451 |
+
"learning_rate": 9.603269973053674e-06,
|
| 2452 |
+
"loss": 0.4148,
|
| 2453 |
+
"step": 177500
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.43,
|
| 2457 |
+
"learning_rate": 9.489476367343966e-06,
|
| 2458 |
+
"loss": 0.4115,
|
| 2459 |
+
"step": 178000
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 2.44,
|
| 2463 |
+
"learning_rate": 9.375682761634259e-06,
|
| 2464 |
+
"loss": 0.4183,
|
| 2465 |
+
"step": 178500
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 2.44,
|
| 2469 |
+
"learning_rate": 9.26188915592455e-06,
|
| 2470 |
+
"loss": 0.4162,
|
| 2471 |
+
"step": 179000
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 2.45,
|
| 2475 |
+
"learning_rate": 9.148095550214843e-06,
|
| 2476 |
+
"loss": 0.411,
|
| 2477 |
+
"step": 179500
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 2.46,
|
| 2481 |
+
"learning_rate": 9.034301944505135e-06,
|
| 2482 |
+
"loss": 0.4122,
|
| 2483 |
+
"step": 180000
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 2.46,
|
| 2487 |
+
"eval_accuracy": 0.90535241823386,
|
| 2488 |
+
"eval_loss": 0.38498830795288086,
|
| 2489 |
+
"eval_runtime": 2156.8626,
|
| 2490 |
+
"eval_samples_per_second": 67.998,
|
| 2491 |
+
"eval_steps_per_second": 17.0,
|
| 2492 |
+
"step": 180000
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 2.46,
|
| 2496 |
+
"learning_rate": 8.920508338795427e-06,
|
| 2497 |
+
"loss": 0.4101,
|
| 2498 |
+
"step": 180500
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 2.47,
|
| 2502 |
+
"learning_rate": 8.80671473308572e-06,
|
| 2503 |
+
"loss": 0.4126,
|
| 2504 |
+
"step": 181000
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 2.48,
|
| 2508 |
+
"learning_rate": 8.69292112737601e-06,
|
| 2509 |
+
"loss": 0.4106,
|
| 2510 |
+
"step": 181500
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.49,
|
| 2514 |
+
"learning_rate": 8.579127521666302e-06,
|
| 2515 |
+
"loss": 0.4138,
|
| 2516 |
+
"step": 182000
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.49,
|
| 2520 |
+
"learning_rate": 8.465333915956596e-06,
|
| 2521 |
+
"loss": 0.413,
|
| 2522 |
+
"step": 182500
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 2.5,
|
| 2526 |
+
"learning_rate": 8.351540310246887e-06,
|
| 2527 |
+
"loss": 0.4105,
|
| 2528 |
+
"step": 183000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 2.51,
|
| 2532 |
+
"learning_rate": 8.237746704537179e-06,
|
| 2533 |
+
"loss": 0.4149,
|
| 2534 |
+
"step": 183500
|
| 2535 |
+
},
|
| 2536 |
+
{
|
| 2537 |
+
"epoch": 2.51,
|
| 2538 |
+
"learning_rate": 8.123953098827471e-06,
|
| 2539 |
+
"loss": 0.4149,
|
| 2540 |
+
"step": 184000
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 2.52,
|
| 2544 |
+
"learning_rate": 8.010159493117763e-06,
|
| 2545 |
+
"loss": 0.41,
|
| 2546 |
+
"step": 184500
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 2.53,
|
| 2550 |
+
"learning_rate": 7.896365887408056e-06,
|
| 2551 |
+
"loss": 0.4065,
|
| 2552 |
+
"step": 185000
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.53,
|
| 2556 |
+
"eval_accuracy": 0.9059730589750316,
|
| 2557 |
+
"eval_loss": 0.3823322355747223,
|
| 2558 |
+
"eval_runtime": 2156.2214,
|
| 2559 |
+
"eval_samples_per_second": 68.019,
|
| 2560 |
+
"eval_steps_per_second": 17.005,
|
| 2561 |
+
"step": 185000
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 2.53,
|
| 2565 |
+
"learning_rate": 7.782572281698348e-06,
|
| 2566 |
+
"loss": 0.4057,
|
| 2567 |
+
"step": 185500
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 2.54,
|
| 2571 |
+
"learning_rate": 7.668778675988638e-06,
|
| 2572 |
+
"loss": 0.4132,
|
| 2573 |
+
"step": 186000
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 2.55,
|
| 2577 |
+
"learning_rate": 7.554985070278932e-06,
|
| 2578 |
+
"loss": 0.4108,
|
| 2579 |
+
"step": 186500
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 2.55,
|
| 2583 |
+
"learning_rate": 7.4411914645692235e-06,
|
| 2584 |
+
"loss": 0.4088,
|
| 2585 |
+
"step": 187000
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 2.56,
|
| 2589 |
+
"learning_rate": 7.327397858859515e-06,
|
| 2590 |
+
"loss": 0.4101,
|
| 2591 |
+
"step": 187500
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 2.57,
|
| 2595 |
+
"learning_rate": 7.213604253149807e-06,
|
| 2596 |
+
"loss": 0.413,
|
| 2597 |
+
"step": 188000
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 2.57,
|
| 2601 |
+
"learning_rate": 7.0998106474401e-06,
|
| 2602 |
+
"loss": 0.4097,
|
| 2603 |
+
"step": 188500
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 2.58,
|
| 2607 |
+
"learning_rate": 6.986017041730391e-06,
|
| 2608 |
+
"loss": 0.4049,
|
| 2609 |
+
"step": 189000
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 2.59,
|
| 2613 |
+
"learning_rate": 6.872223436020684e-06,
|
| 2614 |
+
"loss": 0.4086,
|
| 2615 |
+
"step": 189500
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 2.59,
|
| 2619 |
+
"learning_rate": 6.758429830310975e-06,
|
| 2620 |
+
"loss": 0.4056,
|
| 2621 |
+
"step": 190000
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 2.59,
|
| 2625 |
+
"eval_accuracy": 0.906182177835412,
|
| 2626 |
+
"eval_loss": 0.38130033016204834,
|
| 2627 |
+
"eval_runtime": 2156.4695,
|
| 2628 |
+
"eval_samples_per_second": 68.011,
|
| 2629 |
+
"eval_steps_per_second": 17.003,
|
| 2630 |
+
"step": 190000
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 2.6,
|
| 2634 |
+
"learning_rate": 6.644636224601268e-06,
|
| 2635 |
+
"loss": 0.4059,
|
| 2636 |
+
"step": 190500
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 2.61,
|
| 2640 |
+
"learning_rate": 6.530842618891559e-06,
|
| 2641 |
+
"loss": 0.4041,
|
| 2642 |
+
"step": 191000
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 2.61,
|
| 2646 |
+
"learning_rate": 6.4170490131818516e-06,
|
| 2647 |
+
"loss": 0.4107,
|
| 2648 |
+
"step": 191500
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 2.62,
|
| 2652 |
+
"learning_rate": 6.303255407472143e-06,
|
| 2653 |
+
"loss": 0.4053,
|
| 2654 |
+
"step": 192000
|
| 2655 |
+
},
|
| 2656 |
+
{
|
| 2657 |
+
"epoch": 2.63,
|
| 2658 |
+
"learning_rate": 6.189461801762435e-06,
|
| 2659 |
+
"loss": 0.4107,
|
| 2660 |
+
"step": 192500
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 2.64,
|
| 2664 |
+
"learning_rate": 6.075668196052728e-06,
|
| 2665 |
+
"loss": 0.4038,
|
| 2666 |
+
"step": 193000
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 2.64,
|
| 2670 |
+
"learning_rate": 5.9618745903430195e-06,
|
| 2671 |
+
"loss": 0.41,
|
| 2672 |
+
"step": 193500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 2.65,
|
| 2676 |
+
"learning_rate": 5.848080984633312e-06,
|
| 2677 |
+
"loss": 0.4074,
|
| 2678 |
+
"step": 194000
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 2.66,
|
| 2682 |
+
"learning_rate": 5.734287378923604e-06,
|
| 2683 |
+
"loss": 0.4081,
|
| 2684 |
+
"step": 194500
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 2.66,
|
| 2688 |
+
"learning_rate": 5.620493773213896e-06,
|
| 2689 |
+
"loss": 0.4074,
|
| 2690 |
+
"step": 195000
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 2.66,
|
| 2694 |
+
"eval_accuracy": 0.9065305015342497,
|
| 2695 |
+
"eval_loss": 0.3795691132545471,
|
| 2696 |
+
"eval_runtime": 2156.844,
|
| 2697 |
+
"eval_samples_per_second": 67.999,
|
| 2698 |
+
"eval_steps_per_second": 17.0,
|
| 2699 |
+
"step": 195000
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 2.67,
|
| 2703 |
+
"learning_rate": 5.5067001675041875e-06,
|
| 2704 |
+
"loss": 0.4122,
|
| 2705 |
+
"step": 195500
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 2.68,
|
| 2709 |
+
"learning_rate": 5.39290656179448e-06,
|
| 2710 |
+
"loss": 0.4066,
|
| 2711 |
+
"step": 196000
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 2.68,
|
| 2715 |
+
"learning_rate": 5.279112956084772e-06,
|
| 2716 |
+
"loss": 0.4071,
|
| 2717 |
+
"step": 196500
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 2.69,
|
| 2721 |
+
"learning_rate": 5.165319350375064e-06,
|
| 2722 |
+
"loss": 0.4049,
|
| 2723 |
+
"step": 197000
|
| 2724 |
+
},
|
| 2725 |
+
{
|
| 2726 |
+
"epoch": 2.7,
|
| 2727 |
+
"learning_rate": 5.0515257446653554e-06,
|
| 2728 |
+
"loss": 0.4028,
|
| 2729 |
+
"step": 197500
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 2.7,
|
| 2733 |
+
"learning_rate": 4.9377321389556485e-06,
|
| 2734 |
+
"loss": 0.4064,
|
| 2735 |
+
"step": 198000
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 2.71,
|
| 2739 |
+
"learning_rate": 4.82393853324594e-06,
|
| 2740 |
+
"loss": 0.4057,
|
| 2741 |
+
"step": 198500
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 2.72,
|
| 2745 |
+
"learning_rate": 4.710144927536232e-06,
|
| 2746 |
+
"loss": 0.4029,
|
| 2747 |
+
"step": 199000
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 2.72,
|
| 2751 |
+
"learning_rate": 4.596351321826524e-06,
|
| 2752 |
+
"loss": 0.4068,
|
| 2753 |
+
"step": 199500
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 2.73,
|
| 2757 |
+
"learning_rate": 4.4825577161168164e-06,
|
| 2758 |
+
"loss": 0.4046,
|
| 2759 |
+
"step": 200000
|
| 2760 |
+
},
|
| 2761 |
+
{
|
| 2762 |
+
"epoch": 2.73,
|
| 2763 |
+
"eval_accuracy": 0.9069610695948094,
|
| 2764 |
+
"eval_loss": 0.3776489198207855,
|
| 2765 |
+
"eval_runtime": 2156.5964,
|
| 2766 |
+
"eval_samples_per_second": 68.007,
|
| 2767 |
+
"eval_steps_per_second": 17.002,
|
| 2768 |
+
"step": 200000
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 2.74,
|
| 2772 |
+
"learning_rate": 4.368764110407108e-06,
|
| 2773 |
+
"loss": 0.4016,
|
| 2774 |
+
"step": 200500
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 2.74,
|
| 2778 |
+
"learning_rate": 4.2549705046974e-06,
|
| 2779 |
+
"loss": 0.404,
|
| 2780 |
+
"step": 201000
|
| 2781 |
+
},
|
| 2782 |
+
{
|
| 2783 |
+
"epoch": 2.75,
|
| 2784 |
+
"learning_rate": 4.141176898987692e-06,
|
| 2785 |
+
"loss": 0.4043,
|
| 2786 |
+
"step": 201500
|
| 2787 |
+
},
|
| 2788 |
+
{
|
| 2789 |
+
"epoch": 2.76,
|
| 2790 |
+
"learning_rate": 4.027383293277984e-06,
|
| 2791 |
+
"loss": 0.4043,
|
| 2792 |
+
"step": 202000
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 2.77,
|
| 2796 |
+
"learning_rate": 3.913589687568276e-06,
|
| 2797 |
+
"loss": 0.4008,
|
| 2798 |
+
"step": 202500
|
| 2799 |
+
},
|
| 2800 |
+
{
|
| 2801 |
+
"epoch": 2.77,
|
| 2802 |
+
"learning_rate": 3.7997960818585684e-06,
|
| 2803 |
+
"loss": 0.4055,
|
| 2804 |
+
"step": 203000
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 2.78,
|
| 2808 |
+
"learning_rate": 3.68600247614886e-06,
|
| 2809 |
+
"loss": 0.403,
|
| 2810 |
+
"step": 203500
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 2.79,
|
| 2814 |
+
"learning_rate": 3.5722088704391523e-06,
|
| 2815 |
+
"loss": 0.4005,
|
| 2816 |
+
"step": 204000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 2.79,
|
| 2820 |
+
"learning_rate": 3.458415264729444e-06,
|
| 2821 |
+
"loss": 0.4009,
|
| 2822 |
+
"step": 204500
|
| 2823 |
+
},
|
| 2824 |
+
{
|
| 2825 |
+
"epoch": 2.8,
|
| 2826 |
+
"learning_rate": 3.3446216590197367e-06,
|
| 2827 |
+
"loss": 0.4061,
|
| 2828 |
+
"step": 205000
|
| 2829 |
+
},
|
| 2830 |
+
{
|
| 2831 |
+
"epoch": 2.8,
|
| 2832 |
+
"eval_accuracy": 0.9072532294618356,
|
| 2833 |
+
"eval_loss": 0.3758757412433624,
|
| 2834 |
+
"eval_runtime": 2157.4549,
|
| 2835 |
+
"eval_samples_per_second": 67.98,
|
| 2836 |
+
"eval_steps_per_second": 16.995,
|
| 2837 |
+
"step": 205000
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 2.81,
|
| 2841 |
+
"learning_rate": 3.230828053310028e-06,
|
| 2842 |
+
"loss": 0.4059,
|
| 2843 |
+
"step": 205500
|
| 2844 |
+
},
|
| 2845 |
+
{
|
| 2846 |
+
"epoch": 2.81,
|
| 2847 |
+
"learning_rate": 3.1170344476003207e-06,
|
| 2848 |
+
"loss": 0.4046,
|
| 2849 |
+
"step": 206000
|
| 2850 |
+
},
|
| 2851 |
+
{
|
| 2852 |
+
"epoch": 2.82,
|
| 2853 |
+
"learning_rate": 3.0032408418906125e-06,
|
| 2854 |
+
"loss": 0.4037,
|
| 2855 |
+
"step": 206500
|
| 2856 |
+
},
|
| 2857 |
+
{
|
| 2858 |
+
"epoch": 2.83,
|
| 2859 |
+
"learning_rate": 2.8894472361809047e-06,
|
| 2860 |
+
"loss": 0.4011,
|
| 2861 |
+
"step": 207000
|
| 2862 |
+
},
|
| 2863 |
+
{
|
| 2864 |
+
"epoch": 2.83,
|
| 2865 |
+
"learning_rate": 2.775653630471197e-06,
|
| 2866 |
+
"loss": 0.4027,
|
| 2867 |
+
"step": 207500
|
| 2868 |
+
},
|
| 2869 |
+
{
|
| 2870 |
+
"epoch": 2.84,
|
| 2871 |
+
"learning_rate": 2.6618600247614887e-06,
|
| 2872 |
+
"loss": 0.4054,
|
| 2873 |
+
"step": 208000
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 2.85,
|
| 2877 |
+
"learning_rate": 2.548066419051781e-06,
|
| 2878 |
+
"loss": 0.3995,
|
| 2879 |
+
"step": 208500
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 2.85,
|
| 2883 |
+
"learning_rate": 2.4342728133420727e-06,
|
| 2884 |
+
"loss": 0.4019,
|
| 2885 |
+
"step": 209000
|
| 2886 |
+
},
|
| 2887 |
+
{
|
| 2888 |
+
"epoch": 2.86,
|
| 2889 |
+
"learning_rate": 2.320479207632365e-06,
|
| 2890 |
+
"loss": 0.3978,
|
| 2891 |
+
"step": 209500
|
| 2892 |
+
},
|
| 2893 |
+
{
|
| 2894 |
+
"epoch": 2.87,
|
| 2895 |
+
"learning_rate": 2.206685601922657e-06,
|
| 2896 |
+
"loss": 0.4036,
|
| 2897 |
+
"step": 210000
|
| 2898 |
+
},
|
| 2899 |
+
{
|
| 2900 |
+
"epoch": 2.87,
|
| 2901 |
+
"eval_accuracy": 0.9072046194275737,
|
| 2902 |
+
"eval_loss": 0.3758549690246582,
|
| 2903 |
+
"eval_runtime": 2155.2543,
|
| 2904 |
+
"eval_samples_per_second": 68.049,
|
| 2905 |
+
"eval_steps_per_second": 17.012,
|
| 2906 |
+
"step": 210000
|
| 2907 |
+
},
|
| 2908 |
+
{
|
| 2909 |
+
"epoch": 2.87,
|
| 2910 |
+
"learning_rate": 2.092891996212949e-06,
|
| 2911 |
+
"loss": 0.3992,
|
| 2912 |
+
"step": 210500
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 2.88,
|
| 2916 |
+
"learning_rate": 1.979098390503241e-06,
|
| 2917 |
+
"loss": 0.3968,
|
| 2918 |
+
"step": 211000
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"epoch": 2.89,
|
| 2922 |
+
"learning_rate": 1.865304784793533e-06,
|
| 2923 |
+
"loss": 0.4051,
|
| 2924 |
+
"step": 211500
|
| 2925 |
+
},
|
| 2926 |
+
{
|
| 2927 |
+
"epoch": 2.89,
|
| 2928 |
+
"learning_rate": 1.751511179083825e-06,
|
| 2929 |
+
"loss": 0.4044,
|
| 2930 |
+
"step": 212000
|
| 2931 |
+
},
|
| 2932 |
+
{
|
| 2933 |
+
"epoch": 2.9,
|
| 2934 |
+
"learning_rate": 1.637717573374117e-06,
|
| 2935 |
+
"loss": 0.4017,
|
| 2936 |
+
"step": 212500
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 2.91,
|
| 2940 |
+
"learning_rate": 1.523923967664409e-06,
|
| 2941 |
+
"loss": 0.4009,
|
| 2942 |
+
"step": 213000
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 2.92,
|
| 2946 |
+
"learning_rate": 1.410130361954701e-06,
|
| 2947 |
+
"loss": 0.4043,
|
| 2948 |
+
"step": 213500
|
| 2949 |
+
},
|
| 2950 |
+
{
|
| 2951 |
+
"epoch": 2.92,
|
| 2952 |
+
"learning_rate": 1.2963367562449932e-06,
|
| 2953 |
+
"loss": 0.4054,
|
| 2954 |
+
"step": 214000
|
| 2955 |
+
},
|
| 2956 |
+
{
|
| 2957 |
+
"epoch": 2.93,
|
| 2958 |
+
"learning_rate": 1.1825431505352852e-06,
|
| 2959 |
+
"loss": 0.4044,
|
| 2960 |
+
"step": 214500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 2.94,
|
| 2964 |
+
"learning_rate": 1.0687495448255772e-06,
|
| 2965 |
+
"loss": 0.4003,
|
| 2966 |
+
"step": 215000
|
| 2967 |
+
},
|
| 2968 |
+
{
|
| 2969 |
+
"epoch": 2.94,
|
| 2970 |
+
"eval_accuracy": 0.9076027976570412,
|
| 2971 |
+
"eval_loss": 0.3740881383419037,
|
| 2972 |
+
"eval_runtime": 2155.6745,
|
| 2973 |
+
"eval_samples_per_second": 68.036,
|
| 2974 |
+
"eval_steps_per_second": 17.009,
|
| 2975 |
+
"step": 215000
|
| 2976 |
+
},
|
| 2977 |
+
{
|
| 2978 |
+
"epoch": 2.94,
|
| 2979 |
+
"learning_rate": 9.549559391158691e-07,
|
| 2980 |
+
"loss": 0.4014,
|
| 2981 |
+
"step": 215500
|
| 2982 |
+
},
|
| 2983 |
+
{
|
| 2984 |
+
"epoch": 2.95,
|
| 2985 |
+
"learning_rate": 8.411623334061612e-07,
|
| 2986 |
+
"loss": 0.4,
|
| 2987 |
+
"step": 216000
|
| 2988 |
+
},
|
| 2989 |
+
{
|
| 2990 |
+
"epoch": 2.96,
|
| 2991 |
+
"learning_rate": 7.273687276964533e-07,
|
| 2992 |
+
"loss": 0.4007,
|
| 2993 |
+
"step": 216500
|
| 2994 |
+
},
|
| 2995 |
+
{
|
| 2996 |
+
"epoch": 2.96,
|
| 2997 |
+
"learning_rate": 6.135751219867453e-07,
|
| 2998 |
+
"loss": 0.4017,
|
| 2999 |
+
"step": 217000
|
| 3000 |
+
},
|
| 3001 |
+
{
|
| 3002 |
+
"epoch": 2.97,
|
| 3003 |
+
"learning_rate": 4.997815162770374e-07,
|
| 3004 |
+
"loss": 0.4036,
|
| 3005 |
+
"step": 217500
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 2.98,
|
| 3009 |
+
"learning_rate": 3.859879105673294e-07,
|
| 3010 |
+
"loss": 0.4,
|
| 3011 |
+
"step": 218000
|
| 3012 |
+
},
|
| 3013 |
+
{
|
| 3014 |
+
"epoch": 2.98,
|
| 3015 |
+
"learning_rate": 2.7219430485762144e-07,
|
| 3016 |
+
"loss": 0.3958,
|
| 3017 |
+
"step": 218500
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 2.99,
|
| 3021 |
+
"learning_rate": 1.584006991479135e-07,
|
| 3022 |
+
"loss": 0.4019,
|
| 3023 |
+
"step": 219000
|
| 3024 |
+
},
|
| 3025 |
+
{
|
| 3026 |
+
"epoch": 3.0,
|
| 3027 |
+
"learning_rate": 4.4607093438205524e-08,
|
| 3028 |
+
"loss": 0.405,
|
| 3029 |
+
"step": 219500
|
| 3030 |
+
},
|
| 3031 |
+
{
|
| 3032 |
+
"epoch": 3.0,
|
| 3033 |
+
"step": 219696,
|
| 3034 |
+
"total_flos": 9.251953379141652e+17,
|
| 3035 |
+
"train_loss": 0.5021202932867165,
|
| 3036 |
+
"train_runtime": 238578.5378,
|
| 3037 |
+
"train_samples_per_second": 14.734,
|
| 3038 |
+
"train_steps_per_second": 0.921
|
| 3039 |
+
}
|
| 3040 |
+
],
|
| 3041 |
+
"max_steps": 219696,
|
| 3042 |
+
"num_train_epochs": 3,
|
| 3043 |
+
"total_flos": 9.251953379141652e+17,
|
| 3044 |
+
"trial_name": null,
|
| 3045 |
+
"trial_params": null
|
| 3046 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab2efa678c245014657d818950d6f8e9b6740db06e54e43279e3e90a3f430e04
|
| 3 |
+
size 3899
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|