Uploading checkpoint-96000 for xlm-r - mlt-latn
Browse files- .gitattributes +1 -0
- README.md +67 -0
- config.json +34 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- scheduler.pt +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +3 -0
- tokenizer_config.json +54 -0
- trainer_state.json +3105 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: FacebookAI/xlm-roberta-base
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: mlt-Latn
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# mlt-Latn
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [FacebookAI/xlm-roberta-base](https://huggingface.co/FacebookAI/xlm-roberta-base) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.5093
|
| 22 |
+
- Accuracy: 0.9199
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0001
|
| 42 |
+
- train_batch_size: 16
|
| 43 |
+
- eval_batch_size: 16
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 2
|
| 47 |
+
- total_train_batch_size: 32
|
| 48 |
+
- total_eval_batch_size: 32
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: linear
|
| 51 |
+
- training_steps: 100000
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
### Citation Information
|
| 55 |
+
|
| 56 |
+
If you use this model in your work, please cite the following paper. Additionally, if you require more details on training and performance, refer to the paper:
|
| 57 |
+
|
| 58 |
+
@misc{gurgurov2025smallmodelsbigimpact,
|
| 59 |
+
title={Small Models, Big Impact: Efficient Corpus and Graph-Based Adaptation of Small Multilingual Language Models for Low-Resource Languages},
|
| 60 |
+
author={Daniil Gurgurov and Ivan Vykopal and Josef van Genabith and Simon Ostermann},
|
| 61 |
+
year={2025},
|
| 62 |
+
eprint={2502.10140},
|
| 63 |
+
archivePrefix={arXiv},
|
| 64 |
+
primaryClass={cs.CL},
|
| 65 |
+
url={https://arxiv.org/abs/2502.10140},
|
| 66 |
+
}
|
| 67 |
+
|
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "FacebookAI/xlm-roberta-base",
|
| 3 |
+
"adapters": {
|
| 4 |
+
"adapters": {},
|
| 5 |
+
"config_map": {},
|
| 6 |
+
"fusion_config_map": {},
|
| 7 |
+
"fusions": {}
|
| 8 |
+
},
|
| 9 |
+
"architectures": [
|
| 10 |
+
"XLMRobertaForMaskedLM"
|
| 11 |
+
],
|
| 12 |
+
"attention_probs_dropout_prob": 0.1,
|
| 13 |
+
"bos_token_id": 0,
|
| 14 |
+
"classifier_dropout": null,
|
| 15 |
+
"eos_token_id": 2,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.1,
|
| 18 |
+
"hidden_size": 768,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"layer_norm_eps": 1e-05,
|
| 22 |
+
"max_position_embeddings": 514,
|
| 23 |
+
"model_type": "xlm-roberta",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_hidden_layers": 12,
|
| 26 |
+
"output_past": true,
|
| 27 |
+
"pad_token_id": 1,
|
| 28 |
+
"position_embedding_type": "absolute",
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.45.2",
|
| 31 |
+
"type_vocab_size": 1,
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 250002
|
| 34 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04882f1f8d688e5779ed8ec93baba4fab5500bafd92018b08973949308a0dedd
|
| 3 |
+
size 1113205088
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3483629dd3d7799ddc7fe918a4aed1f97390fa62c6005fd8f185eec42cc471e0
|
| 3 |
+
size 2226530490
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2bba032866d845379385ce6954c9b32fd48faf978cbd6207412f7dd38ea654e5
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a972704d602697451d39b92003a9eed5cccdc09b4abb756aa4b87ef94b20c76
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3eba0883374141a9bbd1c16a5c56813754bd11164c435710350a7a977112431
|
| 3 |
+
size 1064
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8373f9cd3d27591e1924426bcc1c8799bc5a9affc4fc857982c5d66668dd1f41
|
| 3 |
+
size 17082832
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"mask_token": "<mask>",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "<pad>",
|
| 51 |
+
"sep_token": "</s>",
|
| 52 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 53 |
+
"unk_token": "<unk>"
|
| 54 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4847618043422699,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/xlm-r/mlt-Latn/checkpoint-96000",
|
| 4 |
+
"epoch": 43.656207366984994,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 96000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.22737608003638018,
|
| 13 |
+
"grad_norm": 5.168328285217285,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.8057,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.22737608003638018,
|
| 20 |
+
"eval_accuracy": 0.7673858936928836,
|
| 21 |
+
"eval_loss": 1.3882914781570435,
|
| 22 |
+
"eval_runtime": 57.4714,
|
| 23 |
+
"eval_samples_per_second": 265.141,
|
| 24 |
+
"eval_steps_per_second": 8.3,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.45475216007276037,
|
| 29 |
+
"grad_norm": 6.386734485626221,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.2412,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.45475216007276037,
|
| 36 |
+
"eval_accuracy": 0.8023059017621376,
|
| 37 |
+
"eval_loss": 1.1490817070007324,
|
| 38 |
+
"eval_runtime": 57.9662,
|
| 39 |
+
"eval_samples_per_second": 262.877,
|
| 40 |
+
"eval_steps_per_second": 8.229,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.6821282401091405,
|
| 45 |
+
"grad_norm": 3.5194196701049805,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.0971,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.6821282401091405,
|
| 52 |
+
"eval_accuracy": 0.8191701818204763,
|
| 53 |
+
"eval_loss": 1.0539811849594116,
|
| 54 |
+
"eval_runtime": 58.0944,
|
| 55 |
+
"eval_samples_per_second": 262.297,
|
| 56 |
+
"eval_steps_per_second": 8.211,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.9095043201455207,
|
| 61 |
+
"grad_norm": 4.919663906097412,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 1.0083,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.9095043201455207,
|
| 68 |
+
"eval_accuracy": 0.8290244187626605,
|
| 69 |
+
"eval_loss": 0.9915244579315186,
|
| 70 |
+
"eval_runtime": 59.0201,
|
| 71 |
+
"eval_samples_per_second": 258.183,
|
| 72 |
+
"eval_steps_per_second": 8.082,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.1368804001819008,
|
| 77 |
+
"grad_norm": 3.4152417182922363,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 0.9343,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1368804001819008,
|
| 84 |
+
"eval_accuracy": 0.8386819891878221,
|
| 85 |
+
"eval_loss": 0.9384229779243469,
|
| 86 |
+
"eval_runtime": 58.1514,
|
| 87 |
+
"eval_samples_per_second": 262.04,
|
| 88 |
+
"eval_steps_per_second": 8.203,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.364256480218281,
|
| 93 |
+
"grad_norm": 3.325963258743286,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 0.8966,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.364256480218281,
|
| 100 |
+
"eval_accuracy": 0.8441958792162307,
|
| 101 |
+
"eval_loss": 0.8955113887786865,
|
| 102 |
+
"eval_runtime": 59.0626,
|
| 103 |
+
"eval_samples_per_second": 257.997,
|
| 104 |
+
"eval_steps_per_second": 8.076,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.591632560254661,
|
| 109 |
+
"grad_norm": 4.558549880981445,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 0.8609,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.591632560254661,
|
| 116 |
+
"eval_accuracy": 0.8482073460109512,
|
| 117 |
+
"eval_loss": 0.8796423673629761,
|
| 118 |
+
"eval_runtime": 58.1755,
|
| 119 |
+
"eval_samples_per_second": 261.931,
|
| 120 |
+
"eval_steps_per_second": 8.199,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.8190086402910413,
|
| 125 |
+
"grad_norm": 4.228474140167236,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 0.8394,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.8190086402910413,
|
| 132 |
+
"eval_accuracy": 0.8522584062680361,
|
| 133 |
+
"eval_loss": 0.8565191030502319,
|
| 134 |
+
"eval_runtime": 59.0465,
|
| 135 |
+
"eval_samples_per_second": 258.068,
|
| 136 |
+
"eval_steps_per_second": 8.078,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0463847203274215,
|
| 141 |
+
"grad_norm": 4.954216003417969,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 0.8082,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.0463847203274215,
|
| 148 |
+
"eval_accuracy": 0.8560422496354393,
|
| 149 |
+
"eval_loss": 0.8301263451576233,
|
| 150 |
+
"eval_runtime": 59.074,
|
| 151 |
+
"eval_samples_per_second": 257.947,
|
| 152 |
+
"eval_steps_per_second": 8.075,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.2737608003638017,
|
| 157 |
+
"grad_norm": 3.8996498584747314,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 0.7751,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.2737608003638017,
|
| 164 |
+
"eval_accuracy": 0.8575601290381725,
|
| 165 |
+
"eval_loss": 0.8123583793640137,
|
| 166 |
+
"eval_runtime": 59.0078,
|
| 167 |
+
"eval_samples_per_second": 258.237,
|
| 168 |
+
"eval_steps_per_second": 8.084,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.501136880400182,
|
| 173 |
+
"grad_norm": 3.683563709259033,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 0.7746,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.501136880400182,
|
| 180 |
+
"eval_accuracy": 0.8608798596484555,
|
| 181 |
+
"eval_loss": 0.8004751801490784,
|
| 182 |
+
"eval_runtime": 57.9526,
|
| 183 |
+
"eval_samples_per_second": 262.939,
|
| 184 |
+
"eval_steps_per_second": 8.231,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.728512960436562,
|
| 189 |
+
"grad_norm": 3.8639516830444336,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 0.7467,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.728512960436562,
|
| 196 |
+
"eval_accuracy": 0.8630581086069647,
|
| 197 |
+
"eval_loss": 0.7939795255661011,
|
| 198 |
+
"eval_runtime": 59.0043,
|
| 199 |
+
"eval_samples_per_second": 258.252,
|
| 200 |
+
"eval_steps_per_second": 8.084,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.9558890404729423,
|
| 205 |
+
"grad_norm": 3.31400990486145,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 0.7404,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.9558890404729423,
|
| 212 |
+
"eval_accuracy": 0.8657109489879384,
|
| 213 |
+
"eval_loss": 0.7746465802192688,
|
| 214 |
+
"eval_runtime": 58.2316,
|
| 215 |
+
"eval_samples_per_second": 261.679,
|
| 216 |
+
"eval_steps_per_second": 8.191,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 3.1832651205093225,
|
| 221 |
+
"grad_norm": 2.843773603439331,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.7193,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.1832651205093225,
|
| 228 |
+
"eval_accuracy": 0.8669591175218982,
|
| 229 |
+
"eval_loss": 0.7662197947502136,
|
| 230 |
+
"eval_runtime": 58.8578,
|
| 231 |
+
"eval_samples_per_second": 258.895,
|
| 232 |
+
"eval_steps_per_second": 8.104,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.4106412005457027,
|
| 237 |
+
"grad_norm": 3.61684513092041,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.7127,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.4106412005457027,
|
| 244 |
+
"eval_accuracy": 0.8696189271143282,
|
| 245 |
+
"eval_loss": 0.7391215562820435,
|
| 246 |
+
"eval_runtime": 58.0068,
|
| 247 |
+
"eval_samples_per_second": 262.693,
|
| 248 |
+
"eval_steps_per_second": 8.223,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.6380172805820825,
|
| 253 |
+
"grad_norm": 2.982285737991333,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.7077,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.6380172805820825,
|
| 260 |
+
"eval_accuracy": 0.8704082502343816,
|
| 261 |
+
"eval_loss": 0.7502180933952332,
|
| 262 |
+
"eval_runtime": 58.7428,
|
| 263 |
+
"eval_samples_per_second": 259.402,
|
| 264 |
+
"eval_steps_per_second": 8.12,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 3.865393360618463,
|
| 269 |
+
"grad_norm": 3.2284772396087646,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.6954,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.865393360618463,
|
| 276 |
+
"eval_accuracy": 0.8723014320863037,
|
| 277 |
+
"eval_loss": 0.7357130646705627,
|
| 278 |
+
"eval_runtime": 57.9037,
|
| 279 |
+
"eval_samples_per_second": 263.161,
|
| 280 |
+
"eval_steps_per_second": 8.238,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 4.092769440654843,
|
| 285 |
+
"grad_norm": 3.141983985900879,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.6831,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.092769440654843,
|
| 292 |
+
"eval_accuracy": 0.87450734136313,
|
| 293 |
+
"eval_loss": 0.7237355709075928,
|
| 294 |
+
"eval_runtime": 58.8298,
|
| 295 |
+
"eval_samples_per_second": 259.018,
|
| 296 |
+
"eval_steps_per_second": 8.108,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.320145520691224,
|
| 301 |
+
"grad_norm": 3.805156707763672,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.6653,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.320145520691224,
|
| 308 |
+
"eval_accuracy": 0.875826148405435,
|
| 309 |
+
"eval_loss": 0.7154198288917542,
|
| 310 |
+
"eval_runtime": 58.8099,
|
| 311 |
+
"eval_samples_per_second": 259.106,
|
| 312 |
+
"eval_steps_per_second": 8.111,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 4.547521600727603,
|
| 317 |
+
"grad_norm": 2.943392753601074,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.6587,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.547521600727603,
|
| 324 |
+
"eval_accuracy": 0.8750290297659766,
|
| 325 |
+
"eval_loss": 0.7158774733543396,
|
| 326 |
+
"eval_runtime": 58.8052,
|
| 327 |
+
"eval_samples_per_second": 259.127,
|
| 328 |
+
"eval_steps_per_second": 8.112,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 4.774897680763983,
|
| 333 |
+
"grad_norm": 3.1016767024993896,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.657,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.774897680763983,
|
| 340 |
+
"eval_accuracy": 0.876631092564109,
|
| 341 |
+
"eval_loss": 0.7100118398666382,
|
| 342 |
+
"eval_runtime": 57.9757,
|
| 343 |
+
"eval_samples_per_second": 262.834,
|
| 344 |
+
"eval_steps_per_second": 8.228,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.002273760800364,
|
| 349 |
+
"grad_norm": 2.9359586238861084,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.6544,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 5.002273760800364,
|
| 356 |
+
"eval_accuracy": 0.8772724264740639,
|
| 357 |
+
"eval_loss": 0.6987695097923279,
|
| 358 |
+
"eval_runtime": 58.0324,
|
| 359 |
+
"eval_samples_per_second": 262.578,
|
| 360 |
+
"eval_steps_per_second": 8.22,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 5.229649840836744,
|
| 365 |
+
"grad_norm": 3.289794445037842,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.6315,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.229649840836744,
|
| 372 |
+
"eval_accuracy": 0.8785936092935085,
|
| 373 |
+
"eval_loss": 0.7098827362060547,
|
| 374 |
+
"eval_runtime": 58.2042,
|
| 375 |
+
"eval_samples_per_second": 261.802,
|
| 376 |
+
"eval_steps_per_second": 8.195,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 5.457025920873124,
|
| 381 |
+
"grad_norm": 3.1603569984436035,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.6406,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.457025920873124,
|
| 388 |
+
"eval_accuracy": 0.8785365360431241,
|
| 389 |
+
"eval_loss": 0.697100043296814,
|
| 390 |
+
"eval_runtime": 58.956,
|
| 391 |
+
"eval_samples_per_second": 258.464,
|
| 392 |
+
"eval_steps_per_second": 8.091,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 5.684402000909504,
|
| 397 |
+
"grad_norm": 3.9137370586395264,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.6296,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 5.684402000909504,
|
| 404 |
+
"eval_accuracy": 0.8807072007268872,
|
| 405 |
+
"eval_loss": 0.6913357377052307,
|
| 406 |
+
"eval_runtime": 58.1403,
|
| 407 |
+
"eval_samples_per_second": 262.09,
|
| 408 |
+
"eval_steps_per_second": 8.204,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 5.911778080945885,
|
| 413 |
+
"grad_norm": 3.703839063644409,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.626,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.911778080945885,
|
| 420 |
+
"eval_accuracy": 0.8809903070027411,
|
| 421 |
+
"eval_loss": 0.6919081807136536,
|
| 422 |
+
"eval_runtime": 58.1489,
|
| 423 |
+
"eval_samples_per_second": 262.051,
|
| 424 |
+
"eval_steps_per_second": 8.203,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 6.139154160982264,
|
| 429 |
+
"grad_norm": 3.5155723094940186,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.6198,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 6.139154160982264,
|
| 436 |
+
"eval_accuracy": 0.8827777558822371,
|
| 437 |
+
"eval_loss": 0.6714188456535339,
|
| 438 |
+
"eval_runtime": 58.9629,
|
| 439 |
+
"eval_samples_per_second": 258.434,
|
| 440 |
+
"eval_steps_per_second": 8.09,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.366530241018645,
|
| 445 |
+
"grad_norm": 3.117462635040283,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.6076,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.366530241018645,
|
| 452 |
+
"eval_accuracy": 0.8834275999297155,
|
| 453 |
+
"eval_loss": 0.6733196973800659,
|
| 454 |
+
"eval_runtime": 59.0324,
|
| 455 |
+
"eval_samples_per_second": 258.13,
|
| 456 |
+
"eval_steps_per_second": 8.08,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 6.593906321055025,
|
| 461 |
+
"grad_norm": 2.6322078704833984,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.6087,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.593906321055025,
|
| 468 |
+
"eval_accuracy": 0.8831569539581932,
|
| 469 |
+
"eval_loss": 0.6722173690795898,
|
| 470 |
+
"eval_runtime": 58.2146,
|
| 471 |
+
"eval_samples_per_second": 261.755,
|
| 472 |
+
"eval_steps_per_second": 8.194,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 6.8212824010914055,
|
| 477 |
+
"grad_norm": 2.875756025314331,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.6016,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 6.8212824010914055,
|
| 484 |
+
"eval_accuracy": 0.8827436768449041,
|
| 485 |
+
"eval_loss": 0.6779712438583374,
|
| 486 |
+
"eval_runtime": 57.9717,
|
| 487 |
+
"eval_samples_per_second": 262.852,
|
| 488 |
+
"eval_steps_per_second": 8.228,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 7.048658481127785,
|
| 493 |
+
"grad_norm": 2.797990560531616,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.5932,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 7.048658481127785,
|
| 500 |
+
"eval_accuracy": 0.8849467695494039,
|
| 501 |
+
"eval_loss": 0.6560626029968262,
|
| 502 |
+
"eval_runtime": 59.2909,
|
| 503 |
+
"eval_samples_per_second": 257.004,
|
| 504 |
+
"eval_steps_per_second": 8.045,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 7.276034561164166,
|
| 509 |
+
"grad_norm": 3.173975706100464,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.5877,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 7.276034561164166,
|
| 516 |
+
"eval_accuracy": 0.8849921630094044,
|
| 517 |
+
"eval_loss": 0.6600627303123474,
|
| 518 |
+
"eval_runtime": 57.9674,
|
| 519 |
+
"eval_samples_per_second": 262.872,
|
| 520 |
+
"eval_steps_per_second": 8.229,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 7.503410641200546,
|
| 525 |
+
"grad_norm": 2.7644402980804443,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.5909,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 7.503410641200546,
|
| 532 |
+
"eval_accuracy": 0.885632879834746,
|
| 533 |
+
"eval_loss": 0.6634506583213806,
|
| 534 |
+
"eval_runtime": 58.1288,
|
| 535 |
+
"eval_samples_per_second": 262.142,
|
| 536 |
+
"eval_steps_per_second": 8.206,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 7.730786721236926,
|
| 541 |
+
"grad_norm": 3.6716108322143555,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.5848,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 7.730786721236926,
|
| 548 |
+
"eval_accuracy": 0.8870372975442135,
|
| 549 |
+
"eval_loss": 0.6555737257003784,
|
| 550 |
+
"eval_runtime": 58.0593,
|
| 551 |
+
"eval_samples_per_second": 262.456,
|
| 552 |
+
"eval_steps_per_second": 8.216,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 7.958162801273306,
|
| 557 |
+
"grad_norm": 2.635899782180786,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.5806,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 7.958162801273306,
|
| 564 |
+
"eval_accuracy": 0.8872147874120453,
|
| 565 |
+
"eval_loss": 0.6475590467453003,
|
| 566 |
+
"eval_runtime": 58.9688,
|
| 567 |
+
"eval_samples_per_second": 258.408,
|
| 568 |
+
"eval_steps_per_second": 8.089,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.185538881309686,
|
| 573 |
+
"grad_norm": 3.155376434326172,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.5717,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 8.185538881309686,
|
| 580 |
+
"eval_accuracy": 0.887904005764964,
|
| 581 |
+
"eval_loss": 0.6570438146591187,
|
| 582 |
+
"eval_runtime": 58.2291,
|
| 583 |
+
"eval_samples_per_second": 261.69,
|
| 584 |
+
"eval_steps_per_second": 8.192,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.412914961346067,
|
| 589 |
+
"grad_norm": 2.7952346801757812,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.564,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.412914961346067,
|
| 596 |
+
"eval_accuracy": 0.8873371327146311,
|
| 597 |
+
"eval_loss": 0.6420606374740601,
|
| 598 |
+
"eval_runtime": 58.022,
|
| 599 |
+
"eval_samples_per_second": 262.624,
|
| 600 |
+
"eval_steps_per_second": 8.221,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 8.640291041382447,
|
| 605 |
+
"grad_norm": 2.901456832885742,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.5668,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 8.640291041382447,
|
| 612 |
+
"eval_accuracy": 0.8888152172025493,
|
| 613 |
+
"eval_loss": 0.6471173763275146,
|
| 614 |
+
"eval_runtime": 58.0374,
|
| 615 |
+
"eval_samples_per_second": 262.555,
|
| 616 |
+
"eval_steps_per_second": 8.219,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 8.867667121418826,
|
| 621 |
+
"grad_norm": 3.783108949661255,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.5661,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 8.867667121418826,
|
| 628 |
+
"eval_accuracy": 0.8896724765068369,
|
| 629 |
+
"eval_loss": 0.6358206272125244,
|
| 630 |
+
"eval_runtime": 57.9826,
|
| 631 |
+
"eval_samples_per_second": 262.803,
|
| 632 |
+
"eval_steps_per_second": 8.227,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 9.095043201455207,
|
| 637 |
+
"grad_norm": 3.0208489894866943,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.5575,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 9.095043201455207,
|
| 644 |
+
"eval_accuracy": 0.8890150798157612,
|
| 645 |
+
"eval_loss": 0.6520903706550598,
|
| 646 |
+
"eval_runtime": 58.9559,
|
| 647 |
+
"eval_samples_per_second": 258.464,
|
| 648 |
+
"eval_steps_per_second": 8.091,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 9.322419281491587,
|
| 653 |
+
"grad_norm": 2.4719395637512207,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.5563,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 9.322419281491587,
|
| 660 |
+
"eval_accuracy": 0.8897703404100196,
|
| 661 |
+
"eval_loss": 0.6332861185073853,
|
| 662 |
+
"eval_runtime": 58.0586,
|
| 663 |
+
"eval_samples_per_second": 262.459,
|
| 664 |
+
"eval_steps_per_second": 8.216,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 9.549795361527968,
|
| 669 |
+
"grad_norm": 3.9511001110076904,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.5462,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 9.549795361527968,
|
| 676 |
+
"eval_accuracy": 0.8907973350119666,
|
| 677 |
+
"eval_loss": 0.6307789087295532,
|
| 678 |
+
"eval_runtime": 58.9193,
|
| 679 |
+
"eval_samples_per_second": 258.625,
|
| 680 |
+
"eval_steps_per_second": 8.096,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 9.777171441564347,
|
| 685 |
+
"grad_norm": 3.5722222328186035,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.554,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 9.777171441564347,
|
| 692 |
+
"eval_accuracy": 0.892567522812507,
|
| 693 |
+
"eval_loss": 0.6097805500030518,
|
| 694 |
+
"eval_runtime": 58.065,
|
| 695 |
+
"eval_samples_per_second": 262.43,
|
| 696 |
+
"eval_steps_per_second": 8.215,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 10.004547521600728,
|
| 701 |
+
"grad_norm": 2.951775074005127,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.5484,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 10.004547521600728,
|
| 708 |
+
"eval_accuracy": 0.8916424877002649,
|
| 709 |
+
"eval_loss": 0.6175746917724609,
|
| 710 |
+
"eval_runtime": 58.4944,
|
| 711 |
+
"eval_samples_per_second": 260.504,
|
| 712 |
+
"eval_steps_per_second": 8.155,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 10.231923601637108,
|
| 717 |
+
"grad_norm": 3.9985241889953613,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.5443,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 10.231923601637108,
|
| 724 |
+
"eval_accuracy": 0.8924841760921375,
|
| 725 |
+
"eval_loss": 0.6330751180648804,
|
| 726 |
+
"eval_runtime": 58.5296,
|
| 727 |
+
"eval_samples_per_second": 260.347,
|
| 728 |
+
"eval_steps_per_second": 8.15,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 10.459299681673487,
|
| 733 |
+
"grad_norm": 3.2177023887634277,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.5329,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 10.459299681673487,
|
| 740 |
+
"eval_accuracy": 0.8919113818187556,
|
| 741 |
+
"eval_loss": 0.6282722353935242,
|
| 742 |
+
"eval_runtime": 58.5087,
|
| 743 |
+
"eval_samples_per_second": 260.44,
|
| 744 |
+
"eval_steps_per_second": 8.153,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 10.686675761709868,
|
| 749 |
+
"grad_norm": 2.4907150268554688,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.5332,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 10.686675761709868,
|
| 756 |
+
"eval_accuracy": 0.8929225353778606,
|
| 757 |
+
"eval_loss": 0.6218137145042419,
|
| 758 |
+
"eval_runtime": 58.6581,
|
| 759 |
+
"eval_samples_per_second": 259.777,
|
| 760 |
+
"eval_steps_per_second": 8.132,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 10.914051841746248,
|
| 765 |
+
"grad_norm": 2.3024277687072754,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.5323,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 10.914051841746248,
|
| 772 |
+
"eval_accuracy": 0.8931125650111673,
|
| 773 |
+
"eval_loss": 0.6083813309669495,
|
| 774 |
+
"eval_runtime": 57.6897,
|
| 775 |
+
"eval_samples_per_second": 264.137,
|
| 776 |
+
"eval_steps_per_second": 8.268,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 11.141427921782629,
|
| 781 |
+
"grad_norm": 3.06503963470459,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.5239,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 11.141427921782629,
|
| 788 |
+
"eval_accuracy": 0.8939010799940649,
|
| 789 |
+
"eval_loss": 0.637162446975708,
|
| 790 |
+
"eval_runtime": 57.7617,
|
| 791 |
+
"eval_samples_per_second": 263.808,
|
| 792 |
+
"eval_steps_per_second": 8.258,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 11.368804001819008,
|
| 797 |
+
"grad_norm": 2.7696726322174072,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.5261,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 11.368804001819008,
|
| 804 |
+
"eval_accuracy": 0.8937561222232755,
|
| 805 |
+
"eval_loss": 0.6276402473449707,
|
| 806 |
+
"eval_runtime": 57.6676,
|
| 807 |
+
"eval_samples_per_second": 264.239,
|
| 808 |
+
"eval_steps_per_second": 8.272,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 11.596180081855389,
|
| 813 |
+
"grad_norm": 2.751497268676758,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.5175,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 11.596180081855389,
|
| 820 |
+
"eval_accuracy": 0.8951661397774281,
|
| 821 |
+
"eval_loss": 0.6050118207931519,
|
| 822 |
+
"eval_runtime": 57.5452,
|
| 823 |
+
"eval_samples_per_second": 264.8,
|
| 824 |
+
"eval_steps_per_second": 8.289,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 11.82355616189177,
|
| 829 |
+
"grad_norm": 2.554222822189331,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.5262,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 11.82355616189177,
|
| 836 |
+
"eval_accuracy": 0.8952365458244355,
|
| 837 |
+
"eval_loss": 0.6047869324684143,
|
| 838 |
+
"eval_runtime": 58.525,
|
| 839 |
+
"eval_samples_per_second": 260.367,
|
| 840 |
+
"eval_steps_per_second": 8.15,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 12.05093224192815,
|
| 845 |
+
"grad_norm": 2.832977294921875,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.5189,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 12.05093224192815,
|
| 852 |
+
"eval_accuracy": 0.8951059547768438,
|
| 853 |
+
"eval_loss": 0.6053177714347839,
|
| 854 |
+
"eval_runtime": 57.7314,
|
| 855 |
+
"eval_samples_per_second": 263.946,
|
| 856 |
+
"eval_steps_per_second": 8.262,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 12.278308321964529,
|
| 861 |
+
"grad_norm": 2.9270408153533936,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.5162,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 12.278308321964529,
|
| 868 |
+
"eval_accuracy": 0.8956970707551482,
|
| 869 |
+
"eval_loss": 0.600378692150116,
|
| 870 |
+
"eval_runtime": 68.0954,
|
| 871 |
+
"eval_samples_per_second": 223.774,
|
| 872 |
+
"eval_steps_per_second": 7.005,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 12.50568440200091,
|
| 877 |
+
"grad_norm": 4.0453782081604,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.5111,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 12.50568440200091,
|
| 884 |
+
"eval_accuracy": 0.8950593575694928,
|
| 885 |
+
"eval_loss": 0.601889431476593,
|
| 886 |
+
"eval_runtime": 69.1002,
|
| 887 |
+
"eval_samples_per_second": 220.52,
|
| 888 |
+
"eval_steps_per_second": 6.903,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 12.73306048203729,
|
| 893 |
+
"grad_norm": 1.8677308559417725,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.51,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 12.73306048203729,
|
| 900 |
+
"eval_accuracy": 0.8962058115277431,
|
| 901 |
+
"eval_loss": 0.6174491047859192,
|
| 902 |
+
"eval_runtime": 69.7483,
|
| 903 |
+
"eval_samples_per_second": 218.471,
|
| 904 |
+
"eval_steps_per_second": 6.839,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 12.96043656207367,
|
| 909 |
+
"grad_norm": 2.551996946334839,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.5109,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 12.96043656207367,
|
| 916 |
+
"eval_accuracy": 0.8969674492010259,
|
| 917 |
+
"eval_loss": 0.6083965301513672,
|
| 918 |
+
"eval_runtime": 66.6755,
|
| 919 |
+
"eval_samples_per_second": 228.54,
|
| 920 |
+
"eval_steps_per_second": 7.154,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 13.18781264211005,
|
| 925 |
+
"grad_norm": 2.474005699157715,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.502,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 13.18781264211005,
|
| 932 |
+
"eval_accuracy": 0.8963457500841246,
|
| 933 |
+
"eval_loss": 0.6135991215705872,
|
| 934 |
+
"eval_runtime": 67.7791,
|
| 935 |
+
"eval_samples_per_second": 224.818,
|
| 936 |
+
"eval_steps_per_second": 7.038,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 13.41518872214643,
|
| 941 |
+
"grad_norm": 3.0140531063079834,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.5,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 13.41518872214643,
|
| 948 |
+
"eval_accuracy": 0.8979126349984058,
|
| 949 |
+
"eval_loss": 0.6121774911880493,
|
| 950 |
+
"eval_runtime": 69.2043,
|
| 951 |
+
"eval_samples_per_second": 220.189,
|
| 952 |
+
"eval_steps_per_second": 6.893,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 13.642564802182811,
|
| 957 |
+
"grad_norm": 2.9221041202545166,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.5024,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 13.642564802182811,
|
| 964 |
+
"eval_accuracy": 0.8981451519977776,
|
| 965 |
+
"eval_loss": 0.59588623046875,
|
| 966 |
+
"eval_runtime": 68.802,
|
| 967 |
+
"eval_samples_per_second": 221.476,
|
| 968 |
+
"eval_steps_per_second": 6.933,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 13.86994088221919,
|
| 973 |
+
"grad_norm": 3.138761043548584,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.4998,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 13.86994088221919,
|
| 980 |
+
"eval_accuracy": 0.8978226704469172,
|
| 981 |
+
"eval_loss": 0.5944364666938782,
|
| 982 |
+
"eval_runtime": 66.5107,
|
| 983 |
+
"eval_samples_per_second": 229.106,
|
| 984 |
+
"eval_steps_per_second": 7.172,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 14.09731696225557,
|
| 989 |
+
"grad_norm": 2.7592108249664307,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.4854,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 14.09731696225557,
|
| 996 |
+
"eval_accuracy": 0.898367703578009,
|
| 997 |
+
"eval_loss": 0.5859193205833435,
|
| 998 |
+
"eval_runtime": 66.7713,
|
| 999 |
+
"eval_samples_per_second": 228.212,
|
| 1000 |
+
"eval_steps_per_second": 7.144,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 14.324693042291951,
|
| 1005 |
+
"grad_norm": 2.9050989151000977,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.4882,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 14.324693042291951,
|
| 1012 |
+
"eval_accuracy": 0.8975941653774508,
|
| 1013 |
+
"eval_loss": 0.60106360912323,
|
| 1014 |
+
"eval_runtime": 68.9335,
|
| 1015 |
+
"eval_samples_per_second": 221.054,
|
| 1016 |
+
"eval_steps_per_second": 6.92,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 14.552069122328332,
|
| 1021 |
+
"grad_norm": 2.3342740535736084,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.4841,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 14.552069122328332,
|
| 1028 |
+
"eval_accuracy": 0.8990730423650459,
|
| 1029 |
+
"eval_loss": 0.584600567817688,
|
| 1030 |
+
"eval_runtime": 67.2181,
|
| 1031 |
+
"eval_samples_per_second": 226.695,
|
| 1032 |
+
"eval_steps_per_second": 7.096,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 14.77944520236471,
|
| 1037 |
+
"grad_norm": 2.1228647232055664,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.4844,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 14.77944520236471,
|
| 1044 |
+
"eval_accuracy": 0.8990018801499072,
|
| 1045 |
+
"eval_loss": 0.5907680988311768,
|
| 1046 |
+
"eval_runtime": 59.205,
|
| 1047 |
+
"eval_samples_per_second": 257.377,
|
| 1048 |
+
"eval_steps_per_second": 8.057,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 15.006821282401091,
|
| 1053 |
+
"grad_norm": 3.1188831329345703,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.4831,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 15.006821282401091,
|
| 1060 |
+
"eval_accuracy": 0.8995549513632177,
|
| 1061 |
+
"eval_loss": 0.5859436392784119,
|
| 1062 |
+
"eval_runtime": 58.4781,
|
| 1063 |
+
"eval_samples_per_second": 260.576,
|
| 1064 |
+
"eval_steps_per_second": 8.157,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 15.234197362437472,
|
| 1069 |
+
"grad_norm": 3.4289638996124268,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.4783,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 15.234197362437472,
|
| 1076 |
+
"eval_accuracy": 0.8997874968645139,
|
| 1077 |
+
"eval_loss": 0.586015522480011,
|
| 1078 |
+
"eval_runtime": 58.2205,
|
| 1079 |
+
"eval_samples_per_second": 261.729,
|
| 1080 |
+
"eval_steps_per_second": 8.193,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 15.461573442473851,
|
| 1085 |
+
"grad_norm": 2.7651004791259766,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.4752,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 15.461573442473851,
|
| 1092 |
+
"eval_accuracy": 0.9008641609570693,
|
| 1093 |
+
"eval_loss": 0.5896708369255066,
|
| 1094 |
+
"eval_runtime": 59.224,
|
| 1095 |
+
"eval_samples_per_second": 257.294,
|
| 1096 |
+
"eval_steps_per_second": 8.054,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 15.688949522510232,
|
| 1101 |
+
"grad_norm": 2.334015130996704,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.4751,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 15.688949522510232,
|
| 1108 |
+
"eval_accuracy": 0.8998034774235811,
|
| 1109 |
+
"eval_loss": 0.5798426270484924,
|
| 1110 |
+
"eval_runtime": 58.5323,
|
| 1111 |
+
"eval_samples_per_second": 260.335,
|
| 1112 |
+
"eval_steps_per_second": 8.149,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 15.916325602546612,
|
| 1117 |
+
"grad_norm": 1.9268873929977417,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.4781,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 15.916325602546612,
|
| 1124 |
+
"eval_accuracy": 0.900376427009191,
|
| 1125 |
+
"eval_loss": 0.574535071849823,
|
| 1126 |
+
"eval_runtime": 59.3062,
|
| 1127 |
+
"eval_samples_per_second": 256.938,
|
| 1128 |
+
"eval_steps_per_second": 8.043,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 16.143701682582993,
|
| 1133 |
+
"grad_norm": 3.1282005310058594,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.4741,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 16.143701682582993,
|
| 1140 |
+
"eval_accuracy": 0.9008166774124221,
|
| 1141 |
+
"eval_loss": 0.5920200347900391,
|
| 1142 |
+
"eval_runtime": 67.8707,
|
| 1143 |
+
"eval_samples_per_second": 224.515,
|
| 1144 |
+
"eval_steps_per_second": 7.028,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 16.37107776261937,
|
| 1149 |
+
"grad_norm": 2.4809906482696533,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.4764,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 16.37107776261937,
|
| 1156 |
+
"eval_accuracy": 0.9007614317314762,
|
| 1157 |
+
"eval_loss": 0.5653102397918701,
|
| 1158 |
+
"eval_runtime": 67.6973,
|
| 1159 |
+
"eval_samples_per_second": 225.09,
|
| 1160 |
+
"eval_steps_per_second": 7.046,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 16.598453842655754,
|
| 1165 |
+
"grad_norm": 2.460899591445923,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.4693,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 16.598453842655754,
|
| 1172 |
+
"eval_accuracy": 0.9008501214346764,
|
| 1173 |
+
"eval_loss": 0.5879611968994141,
|
| 1174 |
+
"eval_runtime": 66.4301,
|
| 1175 |
+
"eval_samples_per_second": 229.384,
|
| 1176 |
+
"eval_steps_per_second": 7.18,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 16.825829922692133,
|
| 1181 |
+
"grad_norm": 2.389681816101074,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.4638,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 16.825829922692133,
|
| 1188 |
+
"eval_accuracy": 0.9024308529635798,
|
| 1189 |
+
"eval_loss": 0.5856647491455078,
|
| 1190 |
+
"eval_runtime": 66.0584,
|
| 1191 |
+
"eval_samples_per_second": 230.675,
|
| 1192 |
+
"eval_steps_per_second": 7.221,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 17.053206002728512,
|
| 1197 |
+
"grad_norm": 2.5536763668060303,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.4643,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 17.053206002728512,
|
| 1204 |
+
"eval_accuracy": 0.9017984039775945,
|
| 1205 |
+
"eval_loss": 0.5838255882263184,
|
| 1206 |
+
"eval_runtime": 66.4217,
|
| 1207 |
+
"eval_samples_per_second": 229.413,
|
| 1208 |
+
"eval_steps_per_second": 7.181,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 17.280582082764894,
|
| 1213 |
+
"grad_norm": 2.8841841220855713,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.4559,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 17.280582082764894,
|
| 1220 |
+
"eval_accuracy": 0.9016667691546006,
|
| 1221 |
+
"eval_loss": 0.5722949504852295,
|
| 1222 |
+
"eval_runtime": 67.5295,
|
| 1223 |
+
"eval_samples_per_second": 225.65,
|
| 1224 |
+
"eval_steps_per_second": 7.064,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 17.507958162801273,
|
| 1229 |
+
"grad_norm": 2.0763895511627197,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.4649,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 17.507958162801273,
|
| 1236 |
+
"eval_accuracy": 0.9028016315170558,
|
| 1237 |
+
"eval_loss": 0.5597889423370361,
|
| 1238 |
+
"eval_runtime": 68.1673,
|
| 1239 |
+
"eval_samples_per_second": 223.538,
|
| 1240 |
+
"eval_steps_per_second": 6.997,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 17.735334242837652,
|
| 1245 |
+
"grad_norm": 2.7098090648651123,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.4619,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 17.735334242837652,
|
| 1252 |
+
"eval_accuracy": 0.9032880942757541,
|
| 1253 |
+
"eval_loss": 0.5635744333267212,
|
| 1254 |
+
"eval_runtime": 65.7624,
|
| 1255 |
+
"eval_samples_per_second": 231.713,
|
| 1256 |
+
"eval_steps_per_second": 7.253,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 17.962710322874035,
|
| 1261 |
+
"grad_norm": 2.361464023590088,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.4554,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 17.962710322874035,
|
| 1268 |
+
"eval_accuracy": 0.9026038090276955,
|
| 1269 |
+
"eval_loss": 0.5572078227996826,
|
| 1270 |
+
"eval_runtime": 65.3403,
|
| 1271 |
+
"eval_samples_per_second": 233.21,
|
| 1272 |
+
"eval_steps_per_second": 7.3,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 18.190086402910413,
|
| 1277 |
+
"grad_norm": 2.2354624271392822,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.4516,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 18.190086402910413,
|
| 1284 |
+
"eval_accuracy": 0.9029999052162649,
|
| 1285 |
+
"eval_loss": 0.5721431970596313,
|
| 1286 |
+
"eval_runtime": 66.8398,
|
| 1287 |
+
"eval_samples_per_second": 227.978,
|
| 1288 |
+
"eval_steps_per_second": 7.136,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 18.417462482946792,
|
| 1293 |
+
"grad_norm": 2.5873236656188965,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.4488,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 18.417462482946792,
|
| 1300 |
+
"eval_accuracy": 0.9031028868580921,
|
| 1301 |
+
"eval_loss": 0.577458381652832,
|
| 1302 |
+
"eval_runtime": 65.9075,
|
| 1303 |
+
"eval_samples_per_second": 231.203,
|
| 1304 |
+
"eval_steps_per_second": 7.237,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 18.644838562983175,
|
| 1309 |
+
"grad_norm": 2.14237642288208,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.4529,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 18.644838562983175,
|
| 1316 |
+
"eval_accuracy": 0.9034652567356332,
|
| 1317 |
+
"eval_loss": 0.5635027289390564,
|
| 1318 |
+
"eval_runtime": 57.1062,
|
| 1319 |
+
"eval_samples_per_second": 266.836,
|
| 1320 |
+
"eval_steps_per_second": 8.353,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 18.872214643019554,
|
| 1325 |
+
"grad_norm": 2.687331199645996,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.4457,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 18.872214643019554,
|
| 1332 |
+
"eval_accuracy": 0.9038571769197852,
|
| 1333 |
+
"eval_loss": 0.5578777194023132,
|
| 1334 |
+
"eval_runtime": 57.0961,
|
| 1335 |
+
"eval_samples_per_second": 266.883,
|
| 1336 |
+
"eval_steps_per_second": 8.354,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 19.099590723055936,
|
| 1341 |
+
"grad_norm": 2.4610774517059326,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.4456,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 19.099590723055936,
|
| 1348 |
+
"eval_accuracy": 0.9040196805071515,
|
| 1349 |
+
"eval_loss": 0.570646345615387,
|
| 1350 |
+
"eval_runtime": 57.0535,
|
| 1351 |
+
"eval_samples_per_second": 267.083,
|
| 1352 |
+
"eval_steps_per_second": 8.361,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 19.326966803092315,
|
| 1357 |
+
"grad_norm": 2.7358736991882324,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.4439,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 19.326966803092315,
|
| 1364 |
+
"eval_accuracy": 0.9047236917555364,
|
| 1365 |
+
"eval_loss": 0.5603917837142944,
|
| 1366 |
+
"eval_runtime": 57.9311,
|
| 1367 |
+
"eval_samples_per_second": 263.037,
|
| 1368 |
+
"eval_steps_per_second": 8.234,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 19.554342883128694,
|
| 1373 |
+
"grad_norm": 2.6887753009796143,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.4371,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 19.554342883128694,
|
| 1380 |
+
"eval_accuracy": 0.9050488743420761,
|
| 1381 |
+
"eval_loss": 0.5606555938720703,
|
| 1382 |
+
"eval_runtime": 57.2573,
|
| 1383 |
+
"eval_samples_per_second": 266.132,
|
| 1384 |
+
"eval_steps_per_second": 8.331,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 19.781718963165076,
|
| 1389 |
+
"grad_norm": 4.247918128967285,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.4419,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 19.781718963165076,
|
| 1396 |
+
"eval_accuracy": 0.9042590959223571,
|
| 1397 |
+
"eval_loss": 0.5585463643074036,
|
| 1398 |
+
"eval_runtime": 66.8598,
|
| 1399 |
+
"eval_samples_per_second": 227.91,
|
| 1400 |
+
"eval_steps_per_second": 7.134,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 20.009095043201455,
|
| 1405 |
+
"grad_norm": 2.268353223800659,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.4415,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 20.009095043201455,
|
| 1412 |
+
"eval_accuracy": 0.9043696805910753,
|
| 1413 |
+
"eval_loss": 0.5533380508422852,
|
| 1414 |
+
"eval_runtime": 75.8782,
|
| 1415 |
+
"eval_samples_per_second": 200.822,
|
| 1416 |
+
"eval_steps_per_second": 6.286,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 20.236471123237834,
|
| 1421 |
+
"grad_norm": 2.3621208667755127,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.4363,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 20.236471123237834,
|
| 1428 |
+
"eval_accuracy": 0.9052580054316933,
|
| 1429 |
+
"eval_loss": 0.561882734298706,
|
| 1430 |
+
"eval_runtime": 76.0871,
|
| 1431 |
+
"eval_samples_per_second": 200.27,
|
| 1432 |
+
"eval_steps_per_second": 6.269,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 20.463847203274216,
|
| 1437 |
+
"grad_norm": 2.7799289226531982,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.4327,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 20.463847203274216,
|
| 1444 |
+
"eval_accuracy": 0.9050759625981704,
|
| 1445 |
+
"eval_loss": 0.5621650815010071,
|
| 1446 |
+
"eval_runtime": 75.1467,
|
| 1447 |
+
"eval_samples_per_second": 202.777,
|
| 1448 |
+
"eval_steps_per_second": 6.348,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 20.691223283310595,
|
| 1453 |
+
"grad_norm": 2.6320040225982666,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.4341,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 20.691223283310595,
|
| 1460 |
+
"eval_accuracy": 0.90537806159975,
|
| 1461 |
+
"eval_loss": 0.5658455491065979,
|
| 1462 |
+
"eval_runtime": 75.0197,
|
| 1463 |
+
"eval_samples_per_second": 203.12,
|
| 1464 |
+
"eval_steps_per_second": 6.358,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 20.918599363346974,
|
| 1469 |
+
"grad_norm": 2.626370668411255,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.436,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 20.918599363346974,
|
| 1476 |
+
"eval_accuracy": 0.9057144525510582,
|
| 1477 |
+
"eval_loss": 0.5474947094917297,
|
| 1478 |
+
"eval_runtime": 74.8407,
|
| 1479 |
+
"eval_samples_per_second": 203.606,
|
| 1480 |
+
"eval_steps_per_second": 6.374,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 21.145975443383357,
|
| 1485 |
+
"grad_norm": 2.2103912830352783,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.43,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 21.145975443383357,
|
| 1492 |
+
"eval_accuracy": 0.9056834598326479,
|
| 1493 |
+
"eval_loss": 0.5598079562187195,
|
| 1494 |
+
"eval_runtime": 67.154,
|
| 1495 |
+
"eval_samples_per_second": 226.911,
|
| 1496 |
+
"eval_steps_per_second": 7.103,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 21.373351523419736,
|
| 1501 |
+
"grad_norm": 2.3939335346221924,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.4253,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 21.373351523419736,
|
| 1508 |
+
"eval_accuracy": 0.9061305526247623,
|
| 1509 |
+
"eval_loss": 0.5633291602134705,
|
| 1510 |
+
"eval_runtime": 66.5353,
|
| 1511 |
+
"eval_samples_per_second": 229.021,
|
| 1512 |
+
"eval_steps_per_second": 7.169,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 21.600727603456118,
|
| 1517 |
+
"grad_norm": 2.4314322471618652,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.4276,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 21.600727603456118,
|
| 1524 |
+
"eval_accuracy": 0.9062991778333528,
|
| 1525 |
+
"eval_loss": 0.5566443800926208,
|
| 1526 |
+
"eval_runtime": 68.0296,
|
| 1527 |
+
"eval_samples_per_second": 223.991,
|
| 1528 |
+
"eval_steps_per_second": 7.012,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 21.828103683492497,
|
| 1533 |
+
"grad_norm": 2.792711019515991,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.4235,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 21.828103683492497,
|
| 1540 |
+
"eval_accuracy": 0.9063092114708347,
|
| 1541 |
+
"eval_loss": 0.5603668093681335,
|
| 1542 |
+
"eval_runtime": 67.1839,
|
| 1543 |
+
"eval_samples_per_second": 226.81,
|
| 1544 |
+
"eval_steps_per_second": 7.1,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 22.055479763528876,
|
| 1549 |
+
"grad_norm": 3.2149298191070557,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.4293,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 22.055479763528876,
|
| 1556 |
+
"eval_accuracy": 0.9061439320028164,
|
| 1557 |
+
"eval_loss": 0.5632808804512024,
|
| 1558 |
+
"eval_runtime": 55.9195,
|
| 1559 |
+
"eval_samples_per_second": 272.499,
|
| 1560 |
+
"eval_steps_per_second": 8.53,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 22.282855843565258,
|
| 1565 |
+
"grad_norm": 2.1803324222564697,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.4226,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 22.282855843565258,
|
| 1572 |
+
"eval_accuracy": 0.9075216851954307,
|
| 1573 |
+
"eval_loss": 0.5528887510299683,
|
| 1574 |
+
"eval_runtime": 56.7606,
|
| 1575 |
+
"eval_samples_per_second": 268.461,
|
| 1576 |
+
"eval_steps_per_second": 8.404,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 22.510231923601637,
|
| 1581 |
+
"grad_norm": 2.293896436691284,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.4193,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 22.510231923601637,
|
| 1588 |
+
"eval_accuracy": 0.907387491792463,
|
| 1589 |
+
"eval_loss": 0.5482432246208191,
|
| 1590 |
+
"eval_runtime": 57.104,
|
| 1591 |
+
"eval_samples_per_second": 266.846,
|
| 1592 |
+
"eval_steps_per_second": 8.353,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 22.737608003638016,
|
| 1597 |
+
"grad_norm": 2.2139928340911865,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.4158,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 22.737608003638016,
|
| 1604 |
+
"eval_accuracy": 0.9077673279064253,
|
| 1605 |
+
"eval_loss": 0.5403118133544922,
|
| 1606 |
+
"eval_runtime": 55.9516,
|
| 1607 |
+
"eval_samples_per_second": 272.342,
|
| 1608 |
+
"eval_steps_per_second": 8.525,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 22.9649840836744,
|
| 1613 |
+
"grad_norm": 2.241081476211548,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.4174,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 22.9649840836744,
|
| 1620 |
+
"eval_accuracy": 0.9083347677575133,
|
| 1621 |
+
"eval_loss": 0.5309577584266663,
|
| 1622 |
+
"eval_runtime": 56.8789,
|
| 1623 |
+
"eval_samples_per_second": 267.903,
|
| 1624 |
+
"eval_steps_per_second": 8.386,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 23.192360163710777,
|
| 1629 |
+
"grad_norm": 3.3256096839904785,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.4191,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 23.192360163710777,
|
| 1636 |
+
"eval_accuracy": 0.9088304562502367,
|
| 1637 |
+
"eval_loss": 0.5443009734153748,
|
| 1638 |
+
"eval_runtime": 56.8135,
|
| 1639 |
+
"eval_samples_per_second": 268.211,
|
| 1640 |
+
"eval_steps_per_second": 8.396,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 23.419736243747156,
|
| 1645 |
+
"grad_norm": 2.716857433319092,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.4071,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 23.419736243747156,
|
| 1652 |
+
"eval_accuracy": 0.9077237472285563,
|
| 1653 |
+
"eval_loss": 0.5535444617271423,
|
| 1654 |
+
"eval_runtime": 56.7929,
|
| 1655 |
+
"eval_samples_per_second": 268.308,
|
| 1656 |
+
"eval_steps_per_second": 8.399,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 23.64711232378354,
|
| 1661 |
+
"grad_norm": 2.466326951980591,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.4159,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 23.64711232378354,
|
| 1668 |
+
"eval_accuracy": 0.9093382373851798,
|
| 1669 |
+
"eval_loss": 0.5515927076339722,
|
| 1670 |
+
"eval_runtime": 56.8114,
|
| 1671 |
+
"eval_samples_per_second": 268.221,
|
| 1672 |
+
"eval_steps_per_second": 8.396,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 23.874488403819917,
|
| 1677 |
+
"grad_norm": 2.6443376541137695,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.4159,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 23.874488403819917,
|
| 1684 |
+
"eval_accuracy": 0.9080766240994744,
|
| 1685 |
+
"eval_loss": 0.5417291522026062,
|
| 1686 |
+
"eval_runtime": 56.0043,
|
| 1687 |
+
"eval_samples_per_second": 272.086,
|
| 1688 |
+
"eval_steps_per_second": 8.517,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 24.1018644838563,
|
| 1693 |
+
"grad_norm": 2.359405994415283,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.4128,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 24.1018644838563,
|
| 1700 |
+
"eval_accuracy": 0.9082854247369369,
|
| 1701 |
+
"eval_loss": 0.546323299407959,
|
| 1702 |
+
"eval_runtime": 55.971,
|
| 1703 |
+
"eval_samples_per_second": 272.248,
|
| 1704 |
+
"eval_steps_per_second": 8.522,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 24.32924056389268,
|
| 1709 |
+
"grad_norm": 2.0880720615386963,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.4034,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 24.32924056389268,
|
| 1716 |
+
"eval_accuracy": 0.9087918433807937,
|
| 1717 |
+
"eval_loss": 0.5541105270385742,
|
| 1718 |
+
"eval_runtime": 58.891,
|
| 1719 |
+
"eval_samples_per_second": 258.749,
|
| 1720 |
+
"eval_steps_per_second": 8.1,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 24.556616643929058,
|
| 1725 |
+
"grad_norm": 2.977452039718628,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.4051,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 24.556616643929058,
|
| 1732 |
+
"eval_accuracy": 0.908003587059478,
|
| 1733 |
+
"eval_loss": 0.5499656200408936,
|
| 1734 |
+
"eval_runtime": 57.0241,
|
| 1735 |
+
"eval_samples_per_second": 267.22,
|
| 1736 |
+
"eval_steps_per_second": 8.365,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 24.78399272396544,
|
| 1741 |
+
"grad_norm": 2.6565568447113037,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.3973,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 24.78399272396544,
|
| 1748 |
+
"eval_accuracy": 0.9084627466876093,
|
| 1749 |
+
"eval_loss": 0.5408248901367188,
|
| 1750 |
+
"eval_runtime": 56.052,
|
| 1751 |
+
"eval_samples_per_second": 271.855,
|
| 1752 |
+
"eval_steps_per_second": 8.51,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 25.01136880400182,
|
| 1757 |
+
"grad_norm": 2.017199993133545,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.4038,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 25.01136880400182,
|
| 1764 |
+
"eval_accuracy": 0.9091495750162909,
|
| 1765 |
+
"eval_loss": 0.531546413898468,
|
| 1766 |
+
"eval_runtime": 56.6501,
|
| 1767 |
+
"eval_samples_per_second": 268.984,
|
| 1768 |
+
"eval_steps_per_second": 8.42,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 25.238744884038198,
|
| 1773 |
+
"grad_norm": 2.50418758392334,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.3993,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 25.238744884038198,
|
| 1780 |
+
"eval_accuracy": 0.9091283316273341,
|
| 1781 |
+
"eval_loss": 0.5372242331504822,
|
| 1782 |
+
"eval_runtime": 56.2979,
|
| 1783 |
+
"eval_samples_per_second": 270.667,
|
| 1784 |
+
"eval_steps_per_second": 8.473,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 25.46612096407458,
|
| 1789 |
+
"grad_norm": 2.4235289096832275,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.3981,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 25.46612096407458,
|
| 1796 |
+
"eval_accuracy": 0.9090792662970008,
|
| 1797 |
+
"eval_loss": 0.5424542427062988,
|
| 1798 |
+
"eval_runtime": 56.0065,
|
| 1799 |
+
"eval_samples_per_second": 272.075,
|
| 1800 |
+
"eval_steps_per_second": 8.517,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 25.69349704411096,
|
| 1805 |
+
"grad_norm": 2.049229621887207,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.3965,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 25.69349704411096,
|
| 1812 |
+
"eval_accuracy": 0.9091030899558308,
|
| 1813 |
+
"eval_loss": 0.5371273159980774,
|
| 1814 |
+
"eval_runtime": 56.9032,
|
| 1815 |
+
"eval_samples_per_second": 267.788,
|
| 1816 |
+
"eval_steps_per_second": 8.383,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 25.920873124147338,
|
| 1821 |
+
"grad_norm": 3.2039127349853516,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.3956,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 25.920873124147338,
|
| 1828 |
+
"eval_accuracy": 0.9102188871200084,
|
| 1829 |
+
"eval_loss": 0.5293972492218018,
|
| 1830 |
+
"eval_runtime": 56.0727,
|
| 1831 |
+
"eval_samples_per_second": 271.754,
|
| 1832 |
+
"eval_steps_per_second": 8.507,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 26.14824920418372,
|
| 1837 |
+
"grad_norm": 2.5481934547424316,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.3968,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 26.14824920418372,
|
| 1844 |
+
"eval_accuracy": 0.9097068103213896,
|
| 1845 |
+
"eval_loss": 0.5351966619491577,
|
| 1846 |
+
"eval_runtime": 55.9717,
|
| 1847 |
+
"eval_samples_per_second": 272.245,
|
| 1848 |
+
"eval_steps_per_second": 8.522,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 26.3756252842201,
|
| 1853 |
+
"grad_norm": 2.174415111541748,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.3902,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 26.3756252842201,
|
| 1860 |
+
"eval_accuracy": 0.9100280555809778,
|
| 1861 |
+
"eval_loss": 0.5361006855964661,
|
| 1862 |
+
"eval_runtime": 56.0896,
|
| 1863 |
+
"eval_samples_per_second": 271.672,
|
| 1864 |
+
"eval_steps_per_second": 8.504,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 26.60300136425648,
|
| 1869 |
+
"grad_norm": 2.4938621520996094,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.3919,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 26.60300136425648,
|
| 1876 |
+
"eval_accuracy": 0.9107032517948321,
|
| 1877 |
+
"eval_loss": 0.5426139831542969,
|
| 1878 |
+
"eval_runtime": 56.0504,
|
| 1879 |
+
"eval_samples_per_second": 271.863,
|
| 1880 |
+
"eval_steps_per_second": 8.51,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 26.83037744429286,
|
| 1885 |
+
"grad_norm": 2.729896306991577,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.3932,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 26.83037744429286,
|
| 1892 |
+
"eval_accuracy": 0.910556622486772,
|
| 1893 |
+
"eval_loss": 0.5372660756111145,
|
| 1894 |
+
"eval_runtime": 56.0573,
|
| 1895 |
+
"eval_samples_per_second": 271.829,
|
| 1896 |
+
"eval_steps_per_second": 8.509,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 27.05775352432924,
|
| 1901 |
+
"grad_norm": 2.150261163711548,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.3902,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 27.05775352432924,
|
| 1908 |
+
"eval_accuracy": 0.9114104771079771,
|
| 1909 |
+
"eval_loss": 0.5252653360366821,
|
| 1910 |
+
"eval_runtime": 56.1245,
|
| 1911 |
+
"eval_samples_per_second": 271.503,
|
| 1912 |
+
"eval_steps_per_second": 8.499,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 27.285129604365622,
|
| 1917 |
+
"grad_norm": 2.6134257316589355,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.393,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 27.285129604365622,
|
| 1924 |
+
"eval_accuracy": 0.910748714906454,
|
| 1925 |
+
"eval_loss": 0.5393661260604858,
|
| 1926 |
+
"eval_runtime": 56.1182,
|
| 1927 |
+
"eval_samples_per_second": 271.534,
|
| 1928 |
+
"eval_steps_per_second": 8.5,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 27.512505684402,
|
| 1933 |
+
"grad_norm": 3.204314947128296,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.3869,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 27.512505684402,
|
| 1940 |
+
"eval_accuracy": 0.9109071564105236,
|
| 1941 |
+
"eval_loss": 0.5372085571289062,
|
| 1942 |
+
"eval_runtime": 55.938,
|
| 1943 |
+
"eval_samples_per_second": 272.409,
|
| 1944 |
+
"eval_steps_per_second": 8.527,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 27.73988176443838,
|
| 1949 |
+
"grad_norm": 2.1017961502075195,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.3869,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 27.73988176443838,
|
| 1956 |
+
"eval_accuracy": 0.911014457937958,
|
| 1957 |
+
"eval_loss": 0.5260709524154663,
|
| 1958 |
+
"eval_runtime": 56.0593,
|
| 1959 |
+
"eval_samples_per_second": 271.819,
|
| 1960 |
+
"eval_steps_per_second": 8.509,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 27.967257844474762,
|
| 1965 |
+
"grad_norm": 1.7981553077697754,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.3817,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 27.967257844474762,
|
| 1972 |
+
"eval_accuracy": 0.9113206772639144,
|
| 1973 |
+
"eval_loss": 0.532031238079071,
|
| 1974 |
+
"eval_runtime": 56.0692,
|
| 1975 |
+
"eval_samples_per_second": 271.771,
|
| 1976 |
+
"eval_steps_per_second": 8.507,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 28.19463392451114,
|
| 1981 |
+
"grad_norm": 2.4235992431640625,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.3781,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 28.19463392451114,
|
| 1988 |
+
"eval_accuracy": 0.9117986222084103,
|
| 1989 |
+
"eval_loss": 0.5406020283699036,
|
| 1990 |
+
"eval_runtime": 56.0456,
|
| 1991 |
+
"eval_samples_per_second": 271.886,
|
| 1992 |
+
"eval_steps_per_second": 8.511,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 28.42201000454752,
|
| 1997 |
+
"grad_norm": 2.256941795349121,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.3793,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 28.42201000454752,
|
| 2004 |
+
"eval_accuracy": 0.9126116944372022,
|
| 2005 |
+
"eval_loss": 0.5184915661811829,
|
| 2006 |
+
"eval_runtime": 56.0344,
|
| 2007 |
+
"eval_samples_per_second": 271.94,
|
| 2008 |
+
"eval_steps_per_second": 8.513,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 28.649386084583902,
|
| 2013 |
+
"grad_norm": 2.4860892295837402,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.3758,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 28.649386084583902,
|
| 2020 |
+
"eval_accuracy": 0.911845377066606,
|
| 2021 |
+
"eval_loss": 0.5190649628639221,
|
| 2022 |
+
"eval_runtime": 55.9764,
|
| 2023 |
+
"eval_samples_per_second": 272.222,
|
| 2024 |
+
"eval_steps_per_second": 8.521,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 28.87676216462028,
|
| 2029 |
+
"grad_norm": 2.1798510551452637,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.382,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 28.87676216462028,
|
| 2036 |
+
"eval_accuracy": 0.9118922815248525,
|
| 2037 |
+
"eval_loss": 0.5215730667114258,
|
| 2038 |
+
"eval_runtime": 56.0709,
|
| 2039 |
+
"eval_samples_per_second": 271.763,
|
| 2040 |
+
"eval_steps_per_second": 8.507,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 29.104138244656664,
|
| 2045 |
+
"grad_norm": 2.404370069503784,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.3763,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 29.104138244656664,
|
| 2052 |
+
"eval_accuracy": 0.912145037637229,
|
| 2053 |
+
"eval_loss": 0.5312708020210266,
|
| 2054 |
+
"eval_runtime": 56.2338,
|
| 2055 |
+
"eval_samples_per_second": 270.976,
|
| 2056 |
+
"eval_steps_per_second": 8.482,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 29.331514324693043,
|
| 2061 |
+
"grad_norm": 2.4613826274871826,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.3788,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 29.331514324693043,
|
| 2068 |
+
"eval_accuracy": 0.9128715433606646,
|
| 2069 |
+
"eval_loss": 0.5222127437591553,
|
| 2070 |
+
"eval_runtime": 56.8998,
|
| 2071 |
+
"eval_samples_per_second": 267.804,
|
| 2072 |
+
"eval_steps_per_second": 8.383,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 29.55889040472942,
|
| 2077 |
+
"grad_norm": 3.3356547355651855,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.3755,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 29.55889040472942,
|
| 2084 |
+
"eval_accuracy": 0.913125950801356,
|
| 2085 |
+
"eval_loss": 0.5145973563194275,
|
| 2086 |
+
"eval_runtime": 56.8962,
|
| 2087 |
+
"eval_samples_per_second": 267.821,
|
| 2088 |
+
"eval_steps_per_second": 8.384,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 29.786266484765804,
|
| 2093 |
+
"grad_norm": 2.2212953567504883,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.3788,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 29.786266484765804,
|
| 2100 |
+
"eval_accuracy": 0.9125449385052034,
|
| 2101 |
+
"eval_loss": 0.531129002571106,
|
| 2102 |
+
"eval_runtime": 56.1683,
|
| 2103 |
+
"eval_samples_per_second": 271.292,
|
| 2104 |
+
"eval_steps_per_second": 8.492,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 30.013642564802183,
|
| 2109 |
+
"grad_norm": 2.8541479110717773,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.3737,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 30.013642564802183,
|
| 2116 |
+
"eval_accuracy": 0.9125566886698588,
|
| 2117 |
+
"eval_loss": 0.5407569408416748,
|
| 2118 |
+
"eval_runtime": 56.0905,
|
| 2119 |
+
"eval_samples_per_second": 271.668,
|
| 2120 |
+
"eval_steps_per_second": 8.504,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 30.24101864483856,
|
| 2125 |
+
"grad_norm": 2.438603162765503,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.3702,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 30.24101864483856,
|
| 2132 |
+
"eval_accuracy": 0.9129054862362733,
|
| 2133 |
+
"eval_loss": 0.5133882761001587,
|
| 2134 |
+
"eval_runtime": 57.1245,
|
| 2135 |
+
"eval_samples_per_second": 266.751,
|
| 2136 |
+
"eval_steps_per_second": 8.35,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 30.468394724874944,
|
| 2141 |
+
"grad_norm": 2.1143336296081543,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.3729,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 30.468394724874944,
|
| 2148 |
+
"eval_accuracy": 0.9124566054408887,
|
| 2149 |
+
"eval_loss": 0.5418105125427246,
|
| 2150 |
+
"eval_runtime": 56.0088,
|
| 2151 |
+
"eval_samples_per_second": 272.065,
|
| 2152 |
+
"eval_steps_per_second": 8.517,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 30.695770804911323,
|
| 2157 |
+
"grad_norm": 2.317859649658203,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.3662,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 30.695770804911323,
|
| 2164 |
+
"eval_accuracy": 0.9135011845911662,
|
| 2165 |
+
"eval_loss": 0.5446010828018188,
|
| 2166 |
+
"eval_runtime": 56.0915,
|
| 2167 |
+
"eval_samples_per_second": 271.663,
|
| 2168 |
+
"eval_steps_per_second": 8.504,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 30.923146884947702,
|
| 2173 |
+
"grad_norm": 2.983668088912964,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.3647,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 30.923146884947702,
|
| 2180 |
+
"eval_accuracy": 0.9138354207606285,
|
| 2181 |
+
"eval_loss": 0.527925968170166,
|
| 2182 |
+
"eval_runtime": 56.0461,
|
| 2183 |
+
"eval_samples_per_second": 271.883,
|
| 2184 |
+
"eval_steps_per_second": 8.511,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 31.150522964984084,
|
| 2189 |
+
"grad_norm": 2.1927847862243652,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.3683,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 31.150522964984084,
|
| 2196 |
+
"eval_accuracy": 0.9147327240782173,
|
| 2197 |
+
"eval_loss": 0.5225592255592346,
|
| 2198 |
+
"eval_runtime": 57.0493,
|
| 2199 |
+
"eval_samples_per_second": 267.102,
|
| 2200 |
+
"eval_steps_per_second": 8.361,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 31.377899045020463,
|
| 2205 |
+
"grad_norm": 2.345428228378296,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.3628,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 31.377899045020463,
|
| 2212 |
+
"eval_accuracy": 0.913656255193207,
|
| 2213 |
+
"eval_loss": 0.5174685716629028,
|
| 2214 |
+
"eval_runtime": 55.9136,
|
| 2215 |
+
"eval_samples_per_second": 272.528,
|
| 2216 |
+
"eval_steps_per_second": 8.531,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 31.605275125056846,
|
| 2221 |
+
"grad_norm": 2.5066728591918945,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.3651,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 31.605275125056846,
|
| 2228 |
+
"eval_accuracy": 0.9143142595628372,
|
| 2229 |
+
"eval_loss": 0.5336447358131409,
|
| 2230 |
+
"eval_runtime": 56.0707,
|
| 2231 |
+
"eval_samples_per_second": 271.764,
|
| 2232 |
+
"eval_steps_per_second": 8.507,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 31.832651205093224,
|
| 2237 |
+
"grad_norm": 2.5718226432800293,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.3621,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 31.832651205093224,
|
| 2244 |
+
"eval_accuracy": 0.914124758029471,
|
| 2245 |
+
"eval_loss": 0.5136735439300537,
|
| 2246 |
+
"eval_runtime": 56.0417,
|
| 2247 |
+
"eval_samples_per_second": 271.905,
|
| 2248 |
+
"eval_steps_per_second": 8.512,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 32.06002728512961,
|
| 2253 |
+
"grad_norm": 3.0197207927703857,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.3598,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 32.06002728512961,
|
| 2260 |
+
"eval_accuracy": 0.9143737680741543,
|
| 2261 |
+
"eval_loss": 0.5096654295921326,
|
| 2262 |
+
"eval_runtime": 56.0653,
|
| 2263 |
+
"eval_samples_per_second": 271.79,
|
| 2264 |
+
"eval_steps_per_second": 8.508,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 32.287403365165986,
|
| 2269 |
+
"grad_norm": 2.932882785797119,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.3583,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 32.287403365165986,
|
| 2276 |
+
"eval_accuracy": 0.9142125358692816,
|
| 2277 |
+
"eval_loss": 0.5173851251602173,
|
| 2278 |
+
"eval_runtime": 56.0814,
|
| 2279 |
+
"eval_samples_per_second": 271.712,
|
| 2280 |
+
"eval_steps_per_second": 8.505,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 32.514779445202365,
|
| 2285 |
+
"grad_norm": 1.9991718530654907,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.3542,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 32.514779445202365,
|
| 2292 |
+
"eval_accuracy": 0.9147026168203635,
|
| 2293 |
+
"eval_loss": 0.522916316986084,
|
| 2294 |
+
"eval_runtime": 55.8503,
|
| 2295 |
+
"eval_samples_per_second": 272.837,
|
| 2296 |
+
"eval_steps_per_second": 8.541,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 32.74215552523874,
|
| 2301 |
+
"grad_norm": 1.8940651416778564,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.356,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 32.74215552523874,
|
| 2308 |
+
"eval_accuracy": 0.9140601083755969,
|
| 2309 |
+
"eval_loss": 0.5267335176467896,
|
| 2310 |
+
"eval_runtime": 56.0286,
|
| 2311 |
+
"eval_samples_per_second": 271.968,
|
| 2312 |
+
"eval_steps_per_second": 8.514,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 32.96953160527512,
|
| 2317 |
+
"grad_norm": 2.228545665740967,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.3554,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 32.96953160527512,
|
| 2324 |
+
"eval_accuracy": 0.9148385388009793,
|
| 2325 |
+
"eval_loss": 0.518826425075531,
|
| 2326 |
+
"eval_runtime": 57.0798,
|
| 2327 |
+
"eval_samples_per_second": 266.96,
|
| 2328 |
+
"eval_steps_per_second": 8.357,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 33.19690768531151,
|
| 2333 |
+
"grad_norm": 1.9578146934509277,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.3503,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 33.19690768531151,
|
| 2340 |
+
"eval_accuracy": 0.9152538892188704,
|
| 2341 |
+
"eval_loss": 0.5155122876167297,
|
| 2342 |
+
"eval_runtime": 56.9103,
|
| 2343 |
+
"eval_samples_per_second": 267.755,
|
| 2344 |
+
"eval_steps_per_second": 8.382,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 33.42428376534789,
|
| 2349 |
+
"grad_norm": 2.4094908237457275,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.3557,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 33.42428376534789,
|
| 2356 |
+
"eval_accuracy": 0.9150505399718056,
|
| 2357 |
+
"eval_loss": 0.5098891854286194,
|
| 2358 |
+
"eval_runtime": 55.8856,
|
| 2359 |
+
"eval_samples_per_second": 272.664,
|
| 2360 |
+
"eval_steps_per_second": 8.535,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 33.651659845384266,
|
| 2365 |
+
"grad_norm": 3.974923610687256,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.3504,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 33.651659845384266,
|
| 2372 |
+
"eval_accuracy": 0.9151609378276823,
|
| 2373 |
+
"eval_loss": 0.5117126703262329,
|
| 2374 |
+
"eval_runtime": 56.0457,
|
| 2375 |
+
"eval_samples_per_second": 271.885,
|
| 2376 |
+
"eval_steps_per_second": 8.511,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 33.879035925420645,
|
| 2381 |
+
"grad_norm": 2.490255355834961,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.3543,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 33.879035925420645,
|
| 2388 |
+
"eval_accuracy": 0.9146644769193912,
|
| 2389 |
+
"eval_loss": 0.5268692970275879,
|
| 2390 |
+
"eval_runtime": 55.9308,
|
| 2391 |
+
"eval_samples_per_second": 272.444,
|
| 2392 |
+
"eval_steps_per_second": 8.528,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 34.106412005457024,
|
| 2397 |
+
"grad_norm": 2.831305742263794,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.352,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 34.106412005457024,
|
| 2404 |
+
"eval_accuracy": 0.9151447226136769,
|
| 2405 |
+
"eval_loss": 0.5093286037445068,
|
| 2406 |
+
"eval_runtime": 56.9838,
|
| 2407 |
+
"eval_samples_per_second": 267.409,
|
| 2408 |
+
"eval_steps_per_second": 8.371,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 34.3337880854934,
|
| 2413 |
+
"grad_norm": 2.085205554962158,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.3511,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 34.3337880854934,
|
| 2420 |
+
"eval_accuracy": 0.9152598801148045,
|
| 2421 |
+
"eval_loss": 0.5099524259567261,
|
| 2422 |
+
"eval_runtime": 56.0735,
|
| 2423 |
+
"eval_samples_per_second": 271.75,
|
| 2424 |
+
"eval_steps_per_second": 8.507,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 34.56116416552979,
|
| 2429 |
+
"grad_norm": 2.5950026512145996,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.3477,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 34.56116416552979,
|
| 2436 |
+
"eval_accuracy": 0.9159398865939414,
|
| 2437 |
+
"eval_loss": 0.5132637023925781,
|
| 2438 |
+
"eval_runtime": 56.0447,
|
| 2439 |
+
"eval_samples_per_second": 271.89,
|
| 2440 |
+
"eval_steps_per_second": 8.511,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 34.78854024556617,
|
| 2445 |
+
"grad_norm": 3.0509707927703857,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.3487,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 34.78854024556617,
|
| 2452 |
+
"eval_accuracy": 0.9158777972417345,
|
| 2453 |
+
"eval_loss": 0.5157153010368347,
|
| 2454 |
+
"eval_runtime": 56.0477,
|
| 2455 |
+
"eval_samples_per_second": 271.876,
|
| 2456 |
+
"eval_steps_per_second": 8.511,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 35.01591632560255,
|
| 2461 |
+
"grad_norm": 3.29341983795166,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.3432,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 35.01591632560255,
|
| 2468 |
+
"eval_accuracy": 0.9159272981938116,
|
| 2469 |
+
"eval_loss": 0.503484308719635,
|
| 2470 |
+
"eval_runtime": 56.0514,
|
| 2471 |
+
"eval_samples_per_second": 271.857,
|
| 2472 |
+
"eval_steps_per_second": 8.51,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 35.243292405638925,
|
| 2477 |
+
"grad_norm": 2.7035512924194336,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.3468,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 35.243292405638925,
|
| 2484 |
+
"eval_accuracy": 0.9168826617753377,
|
| 2485 |
+
"eval_loss": 0.504048228263855,
|
| 2486 |
+
"eval_runtime": 56.0964,
|
| 2487 |
+
"eval_samples_per_second": 271.639,
|
| 2488 |
+
"eval_steps_per_second": 8.503,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 35.470668485675304,
|
| 2493 |
+
"grad_norm": 4.089804649353027,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.3444,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 35.470668485675304,
|
| 2500 |
+
"eval_accuracy": 0.9165130415874907,
|
| 2501 |
+
"eval_loss": 0.5067149996757507,
|
| 2502 |
+
"eval_runtime": 56.0967,
|
| 2503 |
+
"eval_samples_per_second": 271.638,
|
| 2504 |
+
"eval_steps_per_second": 8.503,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 35.69804456571169,
|
| 2509 |
+
"grad_norm": 2.281663656234741,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.3391,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 35.69804456571169,
|
| 2516 |
+
"eval_accuracy": 0.9166254129979897,
|
| 2517 |
+
"eval_loss": 0.4899181127548218,
|
| 2518 |
+
"eval_runtime": 56.9423,
|
| 2519 |
+
"eval_samples_per_second": 267.604,
|
| 2520 |
+
"eval_steps_per_second": 8.377,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 35.92542064574807,
|
| 2525 |
+
"grad_norm": 2.6281392574310303,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.3395,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 35.92542064574807,
|
| 2532 |
+
"eval_accuracy": 0.9163551690154882,
|
| 2533 |
+
"eval_loss": 0.5145460367202759,
|
| 2534 |
+
"eval_runtime": 56.0529,
|
| 2535 |
+
"eval_samples_per_second": 271.85,
|
| 2536 |
+
"eval_steps_per_second": 8.51,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 36.15279672578445,
|
| 2541 |
+
"grad_norm": 2.2223994731903076,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.3405,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 36.15279672578445,
|
| 2548 |
+
"eval_accuracy": 0.9158962396841802,
|
| 2549 |
+
"eval_loss": 0.51621013879776,
|
| 2550 |
+
"eval_runtime": 56.078,
|
| 2551 |
+
"eval_samples_per_second": 271.729,
|
| 2552 |
+
"eval_steps_per_second": 8.506,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 36.38017280582083,
|
| 2557 |
+
"grad_norm": 2.688448190689087,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.3415,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 36.38017280582083,
|
| 2564 |
+
"eval_accuracy": 0.9164056168752696,
|
| 2565 |
+
"eval_loss": 0.5185515284538269,
|
| 2566 |
+
"eval_runtime": 56.0615,
|
| 2567 |
+
"eval_samples_per_second": 271.809,
|
| 2568 |
+
"eval_steps_per_second": 8.509,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 36.607548885857206,
|
| 2573 |
+
"grad_norm": 1.9620684385299683,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.3375,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 36.607548885857206,
|
| 2580 |
+
"eval_accuracy": 0.9163537188934124,
|
| 2581 |
+
"eval_loss": 0.517730176448822,
|
| 2582 |
+
"eval_runtime": 56.9493,
|
| 2583 |
+
"eval_samples_per_second": 267.571,
|
| 2584 |
+
"eval_steps_per_second": 8.376,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 36.834924965893585,
|
| 2589 |
+
"grad_norm": 2.495645761489868,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.3398,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 36.834924965893585,
|
| 2596 |
+
"eval_accuracy": 0.9171486235060389,
|
| 2597 |
+
"eval_loss": 0.5006797313690186,
|
| 2598 |
+
"eval_runtime": 56.9369,
|
| 2599 |
+
"eval_samples_per_second": 267.63,
|
| 2600 |
+
"eval_steps_per_second": 8.378,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 37.06230104592997,
|
| 2605 |
+
"grad_norm": 2.4040660858154297,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.3393,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 37.06230104592997,
|
| 2612 |
+
"eval_accuracy": 0.9175943590821654,
|
| 2613 |
+
"eval_loss": 0.5150498151779175,
|
| 2614 |
+
"eval_runtime": 56.1005,
|
| 2615 |
+
"eval_samples_per_second": 271.62,
|
| 2616 |
+
"eval_steps_per_second": 8.503,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 37.28967712596635,
|
| 2621 |
+
"grad_norm": 2.7558135986328125,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.3344,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 37.28967712596635,
|
| 2628 |
+
"eval_accuracy": 0.9171837999528608,
|
| 2629 |
+
"eval_loss": 0.5036485195159912,
|
| 2630 |
+
"eval_runtime": 56.078,
|
| 2631 |
+
"eval_samples_per_second": 271.729,
|
| 2632 |
+
"eval_steps_per_second": 8.506,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 37.51705320600273,
|
| 2637 |
+
"grad_norm": 2.468700647354126,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.3352,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 37.51705320600273,
|
| 2644 |
+
"eval_accuracy": 0.9175697679179382,
|
| 2645 |
+
"eval_loss": 0.5030218958854675,
|
| 2646 |
+
"eval_runtime": 56.0873,
|
| 2647 |
+
"eval_samples_per_second": 271.684,
|
| 2648 |
+
"eval_steps_per_second": 8.505,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 37.74442928603911,
|
| 2653 |
+
"grad_norm": 2.5978448390960693,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.3329,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 37.74442928603911,
|
| 2660 |
+
"eval_accuracy": 0.9171495853383843,
|
| 2661 |
+
"eval_loss": 0.5056445598602295,
|
| 2662 |
+
"eval_runtime": 56.0025,
|
| 2663 |
+
"eval_samples_per_second": 272.095,
|
| 2664 |
+
"eval_steps_per_second": 8.517,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 37.971805366075486,
|
| 2669 |
+
"grad_norm": 2.1514439582824707,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.3298,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 37.971805366075486,
|
| 2676 |
+
"eval_accuracy": 0.917162721884158,
|
| 2677 |
+
"eval_loss": 0.5005716681480408,
|
| 2678 |
+
"eval_runtime": 57.1445,
|
| 2679 |
+
"eval_samples_per_second": 266.658,
|
| 2680 |
+
"eval_steps_per_second": 8.347,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 38.19918144611187,
|
| 2685 |
+
"grad_norm": 2.6213526725769043,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.3283,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 38.19918144611187,
|
| 2692 |
+
"eval_accuracy": 0.9176722555185908,
|
| 2693 |
+
"eval_loss": 0.5062026977539062,
|
| 2694 |
+
"eval_runtime": 56.1037,
|
| 2695 |
+
"eval_samples_per_second": 271.604,
|
| 2696 |
+
"eval_steps_per_second": 8.502,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 38.42655752614825,
|
| 2701 |
+
"grad_norm": 2.2421905994415283,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.327,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 38.42655752614825,
|
| 2708 |
+
"eval_accuracy": 0.918665843034967,
|
| 2709 |
+
"eval_loss": 0.4939974844455719,
|
| 2710 |
+
"eval_runtime": 56.9571,
|
| 2711 |
+
"eval_samples_per_second": 267.535,
|
| 2712 |
+
"eval_steps_per_second": 8.375,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 38.65393360618463,
|
| 2717 |
+
"grad_norm": 2.6543004512786865,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.3335,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 38.65393360618463,
|
| 2724 |
+
"eval_accuracy": 0.9179265011087174,
|
| 2725 |
+
"eval_loss": 0.5135884284973145,
|
| 2726 |
+
"eval_runtime": 56.0886,
|
| 2727 |
+
"eval_samples_per_second": 271.677,
|
| 2728 |
+
"eval_steps_per_second": 8.504,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 38.88130968622101,
|
| 2733 |
+
"grad_norm": 2.2833940982818604,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.3323,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 38.88130968622101,
|
| 2740 |
+
"eval_accuracy": 0.9180402792823424,
|
| 2741 |
+
"eval_loss": 0.5044585466384888,
|
| 2742 |
+
"eval_runtime": 56.0736,
|
| 2743 |
+
"eval_samples_per_second": 271.75,
|
| 2744 |
+
"eval_steps_per_second": 8.507,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 39.10868576625739,
|
| 2749 |
+
"grad_norm": 2.157496213912964,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.3323,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 39.10868576625739,
|
| 2756 |
+
"eval_accuracy": 0.9178364524991991,
|
| 2757 |
+
"eval_loss": 0.5072239637374878,
|
| 2758 |
+
"eval_runtime": 56.0862,
|
| 2759 |
+
"eval_samples_per_second": 271.689,
|
| 2760 |
+
"eval_steps_per_second": 8.505,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 39.33606184629377,
|
| 2765 |
+
"grad_norm": 2.467801809310913,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.3209,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 39.33606184629377,
|
| 2772 |
+
"eval_accuracy": 0.9178805584191491,
|
| 2773 |
+
"eval_loss": 0.5000079870223999,
|
| 2774 |
+
"eval_runtime": 56.0688,
|
| 2775 |
+
"eval_samples_per_second": 271.773,
|
| 2776 |
+
"eval_steps_per_second": 8.507,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 39.56343792633015,
|
| 2781 |
+
"grad_norm": 1.764186143875122,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.318,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 39.56343792633015,
|
| 2788 |
+
"eval_accuracy": 0.9177274550237315,
|
| 2789 |
+
"eval_loss": 0.5174301862716675,
|
| 2790 |
+
"eval_runtime": 55.9254,
|
| 2791 |
+
"eval_samples_per_second": 272.47,
|
| 2792 |
+
"eval_steps_per_second": 8.529,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 39.79081400636653,
|
| 2797 |
+
"grad_norm": 1.9439367055892944,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.3304,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 39.79081400636653,
|
| 2804 |
+
"eval_accuracy": 0.9182047509597898,
|
| 2805 |
+
"eval_loss": 0.5170900821685791,
|
| 2806 |
+
"eval_runtime": 56.0765,
|
| 2807 |
+
"eval_samples_per_second": 271.736,
|
| 2808 |
+
"eval_steps_per_second": 8.506,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 40.01819008640291,
|
| 2813 |
+
"grad_norm": 2.7483394145965576,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.3269,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 40.01819008640291,
|
| 2820 |
+
"eval_accuracy": 0.9192084760524332,
|
| 2821 |
+
"eval_loss": 0.5051037669181824,
|
| 2822 |
+
"eval_runtime": 56.982,
|
| 2823 |
+
"eval_samples_per_second": 267.418,
|
| 2824 |
+
"eval_steps_per_second": 8.371,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 40.24556616643929,
|
| 2829 |
+
"grad_norm": 1.9884289503097534,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.3226,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 40.24556616643929,
|
| 2836 |
+
"eval_accuracy": 0.9179925787853008,
|
| 2837 |
+
"eval_loss": 0.4963218867778778,
|
| 2838 |
+
"eval_runtime": 56.9742,
|
| 2839 |
+
"eval_samples_per_second": 267.454,
|
| 2840 |
+
"eval_steps_per_second": 8.372,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 40.47294224647567,
|
| 2845 |
+
"grad_norm": 2.4133260250091553,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.3193,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 40.47294224647567,
|
| 2852 |
+
"eval_accuracy": 0.9186944890989619,
|
| 2853 |
+
"eval_loss": 0.5157626867294312,
|
| 2854 |
+
"eval_runtime": 56.072,
|
| 2855 |
+
"eval_samples_per_second": 271.758,
|
| 2856 |
+
"eval_steps_per_second": 8.507,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 40.700318326512054,
|
| 2861 |
+
"grad_norm": 2.426737070083618,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.3204,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 40.700318326512054,
|
| 2868 |
+
"eval_accuracy": 0.918837528271839,
|
| 2869 |
+
"eval_loss": 0.49108728766441345,
|
| 2870 |
+
"eval_runtime": 56.0958,
|
| 2871 |
+
"eval_samples_per_second": 271.643,
|
| 2872 |
+
"eval_steps_per_second": 8.503,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 40.92769440654843,
|
| 2877 |
+
"grad_norm": 2.0523056983947754,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.3234,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 40.92769440654843,
|
| 2884 |
+
"eval_accuracy": 0.9187695504636845,
|
| 2885 |
+
"eval_loss": 0.497799277305603,
|
| 2886 |
+
"eval_runtime": 56.0896,
|
| 2887 |
+
"eval_samples_per_second": 271.673,
|
| 2888 |
+
"eval_steps_per_second": 8.504,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 41.15507048658481,
|
| 2893 |
+
"grad_norm": 2.7144970893859863,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.3211,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 41.15507048658481,
|
| 2900 |
+
"eval_accuracy": 0.9191388218103149,
|
| 2901 |
+
"eval_loss": 0.4986066222190857,
|
| 2902 |
+
"eval_runtime": 56.0597,
|
| 2903 |
+
"eval_samples_per_second": 271.818,
|
| 2904 |
+
"eval_steps_per_second": 8.509,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 41.38244656662119,
|
| 2909 |
+
"grad_norm": 2.1701807975769043,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.3202,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 41.38244656662119,
|
| 2916 |
+
"eval_accuracy": 0.9192327856840563,
|
| 2917 |
+
"eval_loss": 0.5045046210289001,
|
| 2918 |
+
"eval_runtime": 56.0657,
|
| 2919 |
+
"eval_samples_per_second": 271.788,
|
| 2920 |
+
"eval_steps_per_second": 8.508,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 41.60982264665757,
|
| 2925 |
+
"grad_norm": 1.8716968297958374,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.3178,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 41.60982264665757,
|
| 2932 |
+
"eval_accuracy": 0.918940321981522,
|
| 2933 |
+
"eval_loss": 0.5036594867706299,
|
| 2934 |
+
"eval_runtime": 57.7478,
|
| 2935 |
+
"eval_samples_per_second": 263.872,
|
| 2936 |
+
"eval_steps_per_second": 8.26,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 41.83719872669395,
|
| 2941 |
+
"grad_norm": 2.7744719982147217,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.3181,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 41.83719872669395,
|
| 2948 |
+
"eval_accuracy": 0.9190958139064731,
|
| 2949 |
+
"eval_loss": 0.49405789375305176,
|
| 2950 |
+
"eval_runtime": 55.9547,
|
| 2951 |
+
"eval_samples_per_second": 272.327,
|
| 2952 |
+
"eval_steps_per_second": 8.525,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 42.064574806730334,
|
| 2957 |
+
"grad_norm": 1.9463552236557007,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.3139,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 42.064574806730334,
|
| 2964 |
+
"eval_accuracy": 0.9185921615457032,
|
| 2965 |
+
"eval_loss": 0.5153664946556091,
|
| 2966 |
+
"eval_runtime": 55.9616,
|
| 2967 |
+
"eval_samples_per_second": 272.294,
|
| 2968 |
+
"eval_steps_per_second": 8.524,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 42.29195088676671,
|
| 2973 |
+
"grad_norm": 1.7864753007888794,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.3155,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 42.29195088676671,
|
| 2980 |
+
"eval_accuracy": 0.9193213422174352,
|
| 2981 |
+
"eval_loss": 0.4968840777873993,
|
| 2982 |
+
"eval_runtime": 56.9973,
|
| 2983 |
+
"eval_samples_per_second": 267.346,
|
| 2984 |
+
"eval_steps_per_second": 8.369,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 42.51932696680309,
|
| 2989 |
+
"grad_norm": 2.8307926654815674,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.3225,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 42.51932696680309,
|
| 2996 |
+
"eval_accuracy": 0.9194418625202981,
|
| 2997 |
+
"eval_loss": 0.49802207946777344,
|
| 2998 |
+
"eval_runtime": 57.0104,
|
| 2999 |
+
"eval_samples_per_second": 267.284,
|
| 3000 |
+
"eval_steps_per_second": 8.367,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 42.74670304683947,
|
| 3005 |
+
"grad_norm": 2.3244693279266357,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.3166,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 42.74670304683947,
|
| 3012 |
+
"eval_accuracy": 0.9192593340761245,
|
| 3013 |
+
"eval_loss": 0.5010645985603333,
|
| 3014 |
+
"eval_runtime": 55.9693,
|
| 3015 |
+
"eval_samples_per_second": 272.256,
|
| 3016 |
+
"eval_steps_per_second": 8.523,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 42.97407912687585,
|
| 3021 |
+
"grad_norm": 1.967890977859497,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.3146,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 42.97407912687585,
|
| 3028 |
+
"eval_accuracy": 0.9196343334604102,
|
| 3029 |
+
"eval_loss": 0.48918718099594116,
|
| 3030 |
+
"eval_runtime": 56.094,
|
| 3031 |
+
"eval_samples_per_second": 271.651,
|
| 3032 |
+
"eval_steps_per_second": 8.504,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 43.201455206912236,
|
| 3037 |
+
"grad_norm": 2.064098596572876,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.3152,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 43.201455206912236,
|
| 3044 |
+
"eval_accuracy": 0.9197978271795374,
|
| 3045 |
+
"eval_loss": 0.4905773103237152,
|
| 3046 |
+
"eval_runtime": 56.0802,
|
| 3047 |
+
"eval_samples_per_second": 271.718,
|
| 3048 |
+
"eval_steps_per_second": 8.506,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 43.428831286948615,
|
| 3053 |
+
"grad_norm": 2.1124424934387207,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.3077,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 43.428831286948615,
|
| 3060 |
+
"eval_accuracy": 0.9199429121130404,
|
| 3061 |
+
"eval_loss": 0.4860183894634247,
|
| 3062 |
+
"eval_runtime": 55.9425,
|
| 3063 |
+
"eval_samples_per_second": 272.387,
|
| 3064 |
+
"eval_steps_per_second": 8.527,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 43.656207366984994,
|
| 3069 |
+
"grad_norm": 1.9392900466918945,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.3185,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 43.656207366984994,
|
| 3076 |
+
"eval_accuracy": 0.919319351233956,
|
| 3077 |
+
"eval_loss": 0.4847618043422699,
|
| 3078 |
+
"eval_runtime": 56.9968,
|
| 3079 |
+
"eval_samples_per_second": 267.348,
|
| 3080 |
+
"eval_steps_per_second": 8.369,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
}
|
| 3083 |
+
],
|
| 3084 |
+
"logging_steps": 500,
|
| 3085 |
+
"max_steps": 100000,
|
| 3086 |
+
"num_input_tokens_seen": 0,
|
| 3087 |
+
"num_train_epochs": 46,
|
| 3088 |
+
"save_steps": 500,
|
| 3089 |
+
"stateful_callbacks": {
|
| 3090 |
+
"TrainerControl": {
|
| 3091 |
+
"args": {
|
| 3092 |
+
"should_epoch_stop": false,
|
| 3093 |
+
"should_evaluate": false,
|
| 3094 |
+
"should_log": false,
|
| 3095 |
+
"should_save": true,
|
| 3096 |
+
"should_training_stop": false
|
| 3097 |
+
},
|
| 3098 |
+
"attributes": {}
|
| 3099 |
+
}
|
| 3100 |
+
},
|
| 3101 |
+
"total_flos": 8.105977491866255e+17,
|
| 3102 |
+
"train_batch_size": 16,
|
| 3103 |
+
"trial_name": null,
|
| 3104 |
+
"trial_params": null
|
| 3105 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:305a8438c67b1c5a91d8b9c59fa0edf2277133e80153652b309467a4e520eb38
|
| 3 |
+
size 5176
|