Uploading checkpoint-94000 for mbert - tel-telu
Browse files- README.md +67 -0
- config.json +37 -0
- generation_config.json +5 -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
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +3041 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: google-bert/bert-base-multilingual-cased
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: tel-Telu
|
| 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 |
+
# tel-Telu
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4361
|
| 22 |
+
- Accuracy: 0.9041
|
| 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 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google-bert/bert-base-multilingual-cased",
|
| 3 |
+
"adapters": {
|
| 4 |
+
"adapters": {},
|
| 5 |
+
"config_map": {},
|
| 6 |
+
"fusion_config_map": {},
|
| 7 |
+
"fusions": {}
|
| 8 |
+
},
|
| 9 |
+
"architectures": [
|
| 10 |
+
"BertForMaskedLM"
|
| 11 |
+
],
|
| 12 |
+
"attention_probs_dropout_prob": 0.1,
|
| 13 |
+
"classifier_dropout": null,
|
| 14 |
+
"directionality": "bidi",
|
| 15 |
+
"hidden_act": "gelu",
|
| 16 |
+
"hidden_dropout_prob": 0.1,
|
| 17 |
+
"hidden_size": 768,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"layer_norm_eps": 1e-12,
|
| 21 |
+
"max_position_embeddings": 512,
|
| 22 |
+
"model_type": "bert",
|
| 23 |
+
"num_attention_heads": 12,
|
| 24 |
+
"num_hidden_layers": 12,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"pooler_fc_size": 768,
|
| 27 |
+
"pooler_num_attention_heads": 12,
|
| 28 |
+
"pooler_num_fc_layers": 3,
|
| 29 |
+
"pooler_size_per_head": 128,
|
| 30 |
+
"pooler_type": "first_token_transform",
|
| 31 |
+
"position_embedding_type": "absolute",
|
| 32 |
+
"torch_dtype": "float32",
|
| 33 |
+
"transformers_version": "4.45.2",
|
| 34 |
+
"type_vocab_size": 2,
|
| 35 |
+
"use_cache": true,
|
| 36 |
+
"vocab_size": 119547
|
| 37 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.45.2"
|
| 5 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63a3b779917299250b189e2053c8a56418e2bba6617d8bc9542a4389e3838f26
|
| 3 |
+
size 711921812
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2414cc2fb68a8b8316fcb09f5546c4625e6d9885c68f50c0280e94b06393583f
|
| 3 |
+
size 1423965114
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be10534d036ed11f40952c2c0d876a74ee694ac4e795b73062d3f39bc11cc0fb
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ecc6c080fabce4714a76691b21d5d09caaf5c39a9b2f07ae13bb5b6cab470992
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a66736ccfa8b18f3e896e617a7b701aef56e09c5b5fe36a5735a533fc72e058
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": false,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3041 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.43036332726478577,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/mbert/tel-Telu/checkpoint-94000",
|
| 4 |
+
"epoch": 29.485570890840652,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 94000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.15683814303638646,
|
| 13 |
+
"grad_norm": 3.4005472660064697,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.2189,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.15683814303638646,
|
| 20 |
+
"eval_accuracy": 0.7666090405556419,
|
| 21 |
+
"eval_loss": 1.1007593870162964,
|
| 22 |
+
"eval_runtime": 133.3794,
|
| 23 |
+
"eval_samples_per_second": 133.169,
|
| 24 |
+
"eval_steps_per_second": 4.169,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.3136762860727729,
|
| 29 |
+
"grad_norm": 2.5922815799713135,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.0883,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.3136762860727729,
|
| 36 |
+
"eval_accuracy": 0.7833480365937524,
|
| 37 |
+
"eval_loss": 1.0111207962036133,
|
| 38 |
+
"eval_runtime": 134.5039,
|
| 39 |
+
"eval_samples_per_second": 132.056,
|
| 40 |
+
"eval_steps_per_second": 4.134,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.47051442910915936,
|
| 45 |
+
"grad_norm": 2.5164525508880615,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.0127,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.47051442910915936,
|
| 52 |
+
"eval_accuracy": 0.7933321529990904,
|
| 53 |
+
"eval_loss": 0.9647787809371948,
|
| 54 |
+
"eval_runtime": 133.0834,
|
| 55 |
+
"eval_samples_per_second": 133.465,
|
| 56 |
+
"eval_steps_per_second": 4.178,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6273525721455459,
|
| 61 |
+
"grad_norm": 2.486825704574585,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 0.9756,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.6273525721455459,
|
| 68 |
+
"eval_accuracy": 0.8023192186252056,
|
| 69 |
+
"eval_loss": 0.9206826686859131,
|
| 70 |
+
"eval_runtime": 134.1318,
|
| 71 |
+
"eval_samples_per_second": 132.422,
|
| 72 |
+
"eval_steps_per_second": 4.145,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.7841907151819323,
|
| 77 |
+
"grad_norm": 2.3757729530334473,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 0.9338,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.7841907151819323,
|
| 84 |
+
"eval_accuracy": 0.8079465713675459,
|
| 85 |
+
"eval_loss": 0.8911862373352051,
|
| 86 |
+
"eval_runtime": 133.0509,
|
| 87 |
+
"eval_samples_per_second": 133.498,
|
| 88 |
+
"eval_steps_per_second": 4.179,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.9410288582183187,
|
| 93 |
+
"grad_norm": 2.4007582664489746,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 0.9019,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.9410288582183187,
|
| 100 |
+
"eval_accuracy": 0.8125347880292766,
|
| 101 |
+
"eval_loss": 0.8683245182037354,
|
| 102 |
+
"eval_runtime": 134.0868,
|
| 103 |
+
"eval_samples_per_second": 132.466,
|
| 104 |
+
"eval_steps_per_second": 4.147,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0978670012547052,
|
| 109 |
+
"grad_norm": 2.373544931411743,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 0.8769,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.0978670012547052,
|
| 116 |
+
"eval_accuracy": 0.8171248540815929,
|
| 117 |
+
"eval_loss": 0.8429937958717346,
|
| 118 |
+
"eval_runtime": 133.1539,
|
| 119 |
+
"eval_samples_per_second": 133.395,
|
| 120 |
+
"eval_steps_per_second": 4.176,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.2547051442910915,
|
| 125 |
+
"grad_norm": 2.486654758453369,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 0.8596,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.2547051442910915,
|
| 132 |
+
"eval_accuracy": 0.8205626406536491,
|
| 133 |
+
"eval_loss": 0.8301252126693726,
|
| 134 |
+
"eval_runtime": 134.0793,
|
| 135 |
+
"eval_samples_per_second": 132.474,
|
| 136 |
+
"eval_steps_per_second": 4.147,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.411543287327478,
|
| 141 |
+
"grad_norm": 2.466484546661377,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 0.8362,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.411543287327478,
|
| 148 |
+
"eval_accuracy": 0.8250476250069525,
|
| 149 |
+
"eval_loss": 0.8079118132591248,
|
| 150 |
+
"eval_runtime": 133.973,
|
| 151 |
+
"eval_samples_per_second": 132.579,
|
| 152 |
+
"eval_steps_per_second": 4.15,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.5683814303638646,
|
| 157 |
+
"grad_norm": 2.474668264389038,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 0.8258,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.5683814303638646,
|
| 164 |
+
"eval_accuracy": 0.8274174720205523,
|
| 165 |
+
"eval_loss": 0.797045886516571,
|
| 166 |
+
"eval_runtime": 133.0342,
|
| 167 |
+
"eval_samples_per_second": 133.514,
|
| 168 |
+
"eval_steps_per_second": 4.179,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.725219573400251,
|
| 173 |
+
"grad_norm": 2.4093940258026123,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 0.8118,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.725219573400251,
|
| 180 |
+
"eval_accuracy": 0.8299715943619322,
|
| 181 |
+
"eval_loss": 0.7782195210456848,
|
| 182 |
+
"eval_runtime": 134.0968,
|
| 183 |
+
"eval_samples_per_second": 132.457,
|
| 184 |
+
"eval_steps_per_second": 4.146,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.8820577164366374,
|
| 189 |
+
"grad_norm": 2.195789098739624,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 0.7988,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.8820577164366374,
|
| 196 |
+
"eval_accuracy": 0.8323508690340673,
|
| 197 |
+
"eval_loss": 0.7673488855361938,
|
| 198 |
+
"eval_runtime": 133.9074,
|
| 199 |
+
"eval_samples_per_second": 132.644,
|
| 200 |
+
"eval_steps_per_second": 4.152,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.038895859473024,
|
| 205 |
+
"grad_norm": 2.0892105102539062,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 0.7796,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.038895859473024,
|
| 212 |
+
"eval_accuracy": 0.8355948662536785,
|
| 213 |
+
"eval_loss": 0.7523614168167114,
|
| 214 |
+
"eval_runtime": 134.0514,
|
| 215 |
+
"eval_samples_per_second": 132.501,
|
| 216 |
+
"eval_steps_per_second": 4.148,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.1957340025094103,
|
| 221 |
+
"grad_norm": 2.0777649879455566,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.7692,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.1957340025094103,
|
| 228 |
+
"eval_accuracy": 0.837018381574701,
|
| 229 |
+
"eval_loss": 0.7498050332069397,
|
| 230 |
+
"eval_runtime": 133.1331,
|
| 231 |
+
"eval_samples_per_second": 133.415,
|
| 232 |
+
"eval_steps_per_second": 4.176,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.3525721455457966,
|
| 237 |
+
"grad_norm": 2.439237594604492,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.7607,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.3525721455457966,
|
| 244 |
+
"eval_accuracy": 0.8375820724969026,
|
| 245 |
+
"eval_loss": 0.7395414113998413,
|
| 246 |
+
"eval_runtime": 134.2187,
|
| 247 |
+
"eval_samples_per_second": 132.336,
|
| 248 |
+
"eval_steps_per_second": 4.142,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.509410288582183,
|
| 253 |
+
"grad_norm": 3.9895544052124023,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.7541,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.509410288582183,
|
| 260 |
+
"eval_accuracy": 0.8398097401921676,
|
| 261 |
+
"eval_loss": 0.7369657158851624,
|
| 262 |
+
"eval_runtime": 133.1275,
|
| 263 |
+
"eval_samples_per_second": 133.421,
|
| 264 |
+
"eval_steps_per_second": 4.176,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.6662484316185697,
|
| 269 |
+
"grad_norm": 3.077057123184204,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.7445,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.6662484316185697,
|
| 276 |
+
"eval_accuracy": 0.8424462410611473,
|
| 277 |
+
"eval_loss": 0.716098964214325,
|
| 278 |
+
"eval_runtime": 170.4177,
|
| 279 |
+
"eval_samples_per_second": 104.226,
|
| 280 |
+
"eval_steps_per_second": 3.263,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.823086574654956,
|
| 285 |
+
"grad_norm": 1.7654019594192505,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.7371,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.823086574654956,
|
| 292 |
+
"eval_accuracy": 0.8441994755214394,
|
| 293 |
+
"eval_loss": 0.7167569398880005,
|
| 294 |
+
"eval_runtime": 181.5711,
|
| 295 |
+
"eval_samples_per_second": 97.824,
|
| 296 |
+
"eval_steps_per_second": 3.062,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.9799247176913424,
|
| 301 |
+
"grad_norm": 2.185347318649292,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.7284,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.9799247176913424,
|
| 308 |
+
"eval_accuracy": 0.8454612799934692,
|
| 309 |
+
"eval_loss": 0.7077948451042175,
|
| 310 |
+
"eval_runtime": 162.2289,
|
| 311 |
+
"eval_samples_per_second": 109.487,
|
| 312 |
+
"eval_steps_per_second": 3.427,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.136762860727729,
|
| 317 |
+
"grad_norm": 2.0829274654388428,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.7116,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.136762860727729,
|
| 324 |
+
"eval_accuracy": 0.846357206649314,
|
| 325 |
+
"eval_loss": 0.7018667459487915,
|
| 326 |
+
"eval_runtime": 133.1956,
|
| 327 |
+
"eval_samples_per_second": 133.353,
|
| 328 |
+
"eval_steps_per_second": 4.174,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.2936010037641155,
|
| 333 |
+
"grad_norm": 2.021026134490967,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.7108,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.2936010037641155,
|
| 340 |
+
"eval_accuracy": 0.8479669750098754,
|
| 341 |
+
"eval_loss": 0.6970675587654114,
|
| 342 |
+
"eval_runtime": 133.1992,
|
| 343 |
+
"eval_samples_per_second": 133.349,
|
| 344 |
+
"eval_steps_per_second": 4.174,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.450439146800502,
|
| 349 |
+
"grad_norm": 2.207099199295044,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.7076,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.450439146800502,
|
| 356 |
+
"eval_accuracy": 0.8491771849924095,
|
| 357 |
+
"eval_loss": 0.686812698841095,
|
| 358 |
+
"eval_runtime": 134.185,
|
| 359 |
+
"eval_samples_per_second": 132.37,
|
| 360 |
+
"eval_steps_per_second": 4.144,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 3.6072772898368886,
|
| 365 |
+
"grad_norm": 2.0449960231781006,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.6973,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.6072772898368886,
|
| 372 |
+
"eval_accuracy": 0.8505685393180309,
|
| 373 |
+
"eval_loss": 0.6852427124977112,
|
| 374 |
+
"eval_runtime": 134.2249,
|
| 375 |
+
"eval_samples_per_second": 132.33,
|
| 376 |
+
"eval_steps_per_second": 4.142,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.764115432873275,
|
| 381 |
+
"grad_norm": 2.1044793128967285,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.7008,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.764115432873275,
|
| 388 |
+
"eval_accuracy": 0.8501603065102159,
|
| 389 |
+
"eval_loss": 0.6858202219009399,
|
| 390 |
+
"eval_runtime": 133.0195,
|
| 391 |
+
"eval_samples_per_second": 133.529,
|
| 392 |
+
"eval_steps_per_second": 4.18,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.920953575909661,
|
| 397 |
+
"grad_norm": 1.9273678064346313,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.6889,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.920953575909661,
|
| 404 |
+
"eval_accuracy": 0.8533251475354033,
|
| 405 |
+
"eval_loss": 0.6630256772041321,
|
| 406 |
+
"eval_runtime": 134.0061,
|
| 407 |
+
"eval_samples_per_second": 132.546,
|
| 408 |
+
"eval_steps_per_second": 4.149,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.077791718946048,
|
| 413 |
+
"grad_norm": 2.1206748485565186,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.6792,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.077791718946048,
|
| 420 |
+
"eval_accuracy": 0.8537328032859233,
|
| 421 |
+
"eval_loss": 0.6658462285995483,
|
| 422 |
+
"eval_runtime": 133.2034,
|
| 423 |
+
"eval_samples_per_second": 133.345,
|
| 424 |
+
"eval_steps_per_second": 4.174,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 4.234629861982434,
|
| 429 |
+
"grad_norm": 1.9410268068313599,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.6718,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 4.234629861982434,
|
| 436 |
+
"eval_accuracy": 0.8547022142485862,
|
| 437 |
+
"eval_loss": 0.6640264391899109,
|
| 438 |
+
"eval_runtime": 134.1575,
|
| 439 |
+
"eval_samples_per_second": 132.397,
|
| 440 |
+
"eval_steps_per_second": 4.144,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.391468005018821,
|
| 445 |
+
"grad_norm": 2.1022086143493652,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.6691,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.391468005018821,
|
| 452 |
+
"eval_accuracy": 0.8557769945974418,
|
| 453 |
+
"eval_loss": 0.6567473411560059,
|
| 454 |
+
"eval_runtime": 133.1176,
|
| 455 |
+
"eval_samples_per_second": 133.431,
|
| 456 |
+
"eval_steps_per_second": 4.177,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 4.548306148055207,
|
| 461 |
+
"grad_norm": 2.805241346359253,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.6665,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.548306148055207,
|
| 468 |
+
"eval_accuracy": 0.85596875,
|
| 469 |
+
"eval_loss": 0.6552355289459229,
|
| 470 |
+
"eval_runtime": 134.0623,
|
| 471 |
+
"eval_samples_per_second": 132.491,
|
| 472 |
+
"eval_steps_per_second": 4.147,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 4.705144291091593,
|
| 477 |
+
"grad_norm": 1.8411997556686401,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.6662,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 4.705144291091593,
|
| 484 |
+
"eval_accuracy": 0.8577343403548875,
|
| 485 |
+
"eval_loss": 0.6424122452735901,
|
| 486 |
+
"eval_runtime": 134.198,
|
| 487 |
+
"eval_samples_per_second": 132.357,
|
| 488 |
+
"eval_steps_per_second": 4.143,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.86198243412798,
|
| 493 |
+
"grad_norm": 2.003784656524658,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.6634,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 4.86198243412798,
|
| 500 |
+
"eval_accuracy": 0.8583665570097477,
|
| 501 |
+
"eval_loss": 0.6462827324867249,
|
| 502 |
+
"eval_runtime": 133.215,
|
| 503 |
+
"eval_samples_per_second": 133.333,
|
| 504 |
+
"eval_steps_per_second": 4.174,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.018820577164367,
|
| 509 |
+
"grad_norm": 2.1657097339630127,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.6523,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.018820577164367,
|
| 516 |
+
"eval_accuracy": 0.8574493938713222,
|
| 517 |
+
"eval_loss": 0.6461288928985596,
|
| 518 |
+
"eval_runtime": 134.1965,
|
| 519 |
+
"eval_samples_per_second": 132.358,
|
| 520 |
+
"eval_steps_per_second": 4.143,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 5.175658720200753,
|
| 525 |
+
"grad_norm": 1.941800594329834,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.6402,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 5.175658720200753,
|
| 532 |
+
"eval_accuracy": 0.859093011891753,
|
| 533 |
+
"eval_loss": 0.644975483417511,
|
| 534 |
+
"eval_runtime": 134.1941,
|
| 535 |
+
"eval_samples_per_second": 132.36,
|
| 536 |
+
"eval_steps_per_second": 4.143,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.3324968632371395,
|
| 541 |
+
"grad_norm": 2.066218137741089,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.6462,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 5.3324968632371395,
|
| 548 |
+
"eval_accuracy": 0.859761200619116,
|
| 549 |
+
"eval_loss": 0.6398352384567261,
|
| 550 |
+
"eval_runtime": 134.1771,
|
| 551 |
+
"eval_samples_per_second": 132.377,
|
| 552 |
+
"eval_steps_per_second": 4.144,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 5.489335006273525,
|
| 557 |
+
"grad_norm": 2.111569881439209,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.6367,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.489335006273525,
|
| 564 |
+
"eval_accuracy": 0.8611538438178138,
|
| 565 |
+
"eval_loss": 0.6319767832756042,
|
| 566 |
+
"eval_runtime": 134.1542,
|
| 567 |
+
"eval_samples_per_second": 132.4,
|
| 568 |
+
"eval_steps_per_second": 4.144,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 5.646173149309912,
|
| 573 |
+
"grad_norm": 2.1733057498931885,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.6381,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 5.646173149309912,
|
| 580 |
+
"eval_accuracy": 0.8611586012500368,
|
| 581 |
+
"eval_loss": 0.6293346285820007,
|
| 582 |
+
"eval_runtime": 224.1026,
|
| 583 |
+
"eval_samples_per_second": 79.258,
|
| 584 |
+
"eval_steps_per_second": 2.481,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.803011292346299,
|
| 589 |
+
"grad_norm": 2.059597969055176,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.6351,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 5.803011292346299,
|
| 596 |
+
"eval_accuracy": 0.8625421914089738,
|
| 597 |
+
"eval_loss": 0.6240249872207642,
|
| 598 |
+
"eval_runtime": 134.2326,
|
| 599 |
+
"eval_samples_per_second": 132.323,
|
| 600 |
+
"eval_steps_per_second": 4.142,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 5.959849435382685,
|
| 605 |
+
"grad_norm": 2.409364938735962,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.6317,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 5.959849435382685,
|
| 612 |
+
"eval_accuracy": 0.8630394224283049,
|
| 613 |
+
"eval_loss": 0.6228821277618408,
|
| 614 |
+
"eval_runtime": 134.2427,
|
| 615 |
+
"eval_samples_per_second": 132.313,
|
| 616 |
+
"eval_steps_per_second": 4.142,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 6.1166875784190715,
|
| 621 |
+
"grad_norm": 1.8734099864959717,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.6218,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.1166875784190715,
|
| 628 |
+
"eval_accuracy": 0.8640943075019144,
|
| 629 |
+
"eval_loss": 0.6216642260551453,
|
| 630 |
+
"eval_runtime": 133.3017,
|
| 631 |
+
"eval_samples_per_second": 133.247,
|
| 632 |
+
"eval_steps_per_second": 4.171,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.273525721455458,
|
| 637 |
+
"grad_norm": 2.096306562423706,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.6286,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.273525721455458,
|
| 644 |
+
"eval_accuracy": 0.8642988371346637,
|
| 645 |
+
"eval_loss": 0.6176745891571045,
|
| 646 |
+
"eval_runtime": 135.0688,
|
| 647 |
+
"eval_samples_per_second": 131.503,
|
| 648 |
+
"eval_steps_per_second": 4.116,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 6.430363864491844,
|
| 653 |
+
"grad_norm": 2.144573450088501,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.6161,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 6.430363864491844,
|
| 660 |
+
"eval_accuracy": 0.8648931740759703,
|
| 661 |
+
"eval_loss": 0.6133354306221008,
|
| 662 |
+
"eval_runtime": 135.1576,
|
| 663 |
+
"eval_samples_per_second": 131.417,
|
| 664 |
+
"eval_steps_per_second": 4.114,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.587202007528231,
|
| 669 |
+
"grad_norm": 1.8558435440063477,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.6165,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.587202007528231,
|
| 676 |
+
"eval_accuracy": 0.8658149990226527,
|
| 677 |
+
"eval_loss": 0.6120994091033936,
|
| 678 |
+
"eval_runtime": 134.814,
|
| 679 |
+
"eval_samples_per_second": 131.752,
|
| 680 |
+
"eval_steps_per_second": 4.124,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.744040150564618,
|
| 685 |
+
"grad_norm": 2.032679319381714,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.6077,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 6.744040150564618,
|
| 692 |
+
"eval_accuracy": 0.8662199269822448,
|
| 693 |
+
"eval_loss": 0.6122332215309143,
|
| 694 |
+
"eval_runtime": 133.8975,
|
| 695 |
+
"eval_samples_per_second": 132.654,
|
| 696 |
+
"eval_steps_per_second": 4.152,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 6.900878293601004,
|
| 701 |
+
"grad_norm": 1.9026424884796143,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.6091,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 6.900878293601004,
|
| 708 |
+
"eval_accuracy": 0.8676860945323992,
|
| 709 |
+
"eval_loss": 0.6020417213439941,
|
| 710 |
+
"eval_runtime": 134.7613,
|
| 711 |
+
"eval_samples_per_second": 131.803,
|
| 712 |
+
"eval_steps_per_second": 4.126,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 7.05771643663739,
|
| 717 |
+
"grad_norm": 1.8703612089157104,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.5995,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 7.05771643663739,
|
| 724 |
+
"eval_accuracy": 0.867634858365159,
|
| 725 |
+
"eval_loss": 0.5994914770126343,
|
| 726 |
+
"eval_runtime": 135.1873,
|
| 727 |
+
"eval_samples_per_second": 131.388,
|
| 728 |
+
"eval_steps_per_second": 4.113,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.214554579673777,
|
| 733 |
+
"grad_norm": 1.8793346881866455,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.6026,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 7.214554579673777,
|
| 740 |
+
"eval_accuracy": 0.8685173783196312,
|
| 741 |
+
"eval_loss": 0.6000118851661682,
|
| 742 |
+
"eval_runtime": 134.1891,
|
| 743 |
+
"eval_samples_per_second": 132.365,
|
| 744 |
+
"eval_steps_per_second": 4.143,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 7.371392722710163,
|
| 749 |
+
"grad_norm": 2.028074264526367,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.5954,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 7.371392722710163,
|
| 756 |
+
"eval_accuracy": 0.8693066916967459,
|
| 757 |
+
"eval_loss": 0.5940452814102173,
|
| 758 |
+
"eval_runtime": 135.085,
|
| 759 |
+
"eval_samples_per_second": 131.488,
|
| 760 |
+
"eval_steps_per_second": 4.116,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 7.52823086574655,
|
| 765 |
+
"grad_norm": 1.83831787109375,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.5943,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 7.52823086574655,
|
| 772 |
+
"eval_accuracy": 0.8691108334634806,
|
| 773 |
+
"eval_loss": 0.5954667925834656,
|
| 774 |
+
"eval_runtime": 135.1088,
|
| 775 |
+
"eval_samples_per_second": 131.464,
|
| 776 |
+
"eval_steps_per_second": 4.115,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 7.685069008782936,
|
| 781 |
+
"grad_norm": 1.942333459854126,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.593,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 7.685069008782936,
|
| 788 |
+
"eval_accuracy": 0.8699215271593945,
|
| 789 |
+
"eval_loss": 0.5896523594856262,
|
| 790 |
+
"eval_runtime": 134.2212,
|
| 791 |
+
"eval_samples_per_second": 132.334,
|
| 792 |
+
"eval_steps_per_second": 4.142,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 7.841907151819322,
|
| 797 |
+
"grad_norm": 2.06683611869812,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.5913,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 7.841907151819322,
|
| 804 |
+
"eval_accuracy": 0.8699148812288157,
|
| 805 |
+
"eval_loss": 0.5872038006782532,
|
| 806 |
+
"eval_runtime": 134.9927,
|
| 807 |
+
"eval_samples_per_second": 131.577,
|
| 808 |
+
"eval_steps_per_second": 4.119,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 7.998745294855709,
|
| 813 |
+
"grad_norm": 1.7606611251831055,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.5856,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 7.998745294855709,
|
| 820 |
+
"eval_accuracy": 0.8713286882993854,
|
| 821 |
+
"eval_loss": 0.5867298245429993,
|
| 822 |
+
"eval_runtime": 135.0166,
|
| 823 |
+
"eval_samples_per_second": 131.554,
|
| 824 |
+
"eval_steps_per_second": 4.118,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.155583437892096,
|
| 829 |
+
"grad_norm": 1.9274616241455078,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.5832,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 8.155583437892096,
|
| 836 |
+
"eval_accuracy": 0.8715261899700203,
|
| 837 |
+
"eval_loss": 0.5882822275161743,
|
| 838 |
+
"eval_runtime": 134.0587,
|
| 839 |
+
"eval_samples_per_second": 132.494,
|
| 840 |
+
"eval_steps_per_second": 4.147,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 8.312421580928483,
|
| 845 |
+
"grad_norm": 1.7970761060714722,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.5808,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 8.312421580928483,
|
| 852 |
+
"eval_accuracy": 0.8708605275290493,
|
| 853 |
+
"eval_loss": 0.5834769606590271,
|
| 854 |
+
"eval_runtime": 138.0709,
|
| 855 |
+
"eval_samples_per_second": 128.644,
|
| 856 |
+
"eval_steps_per_second": 4.027,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 8.469259723964868,
|
| 861 |
+
"grad_norm": 1.8796378374099731,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.5814,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 8.469259723964868,
|
| 868 |
+
"eval_accuracy": 0.8719365756329153,
|
| 869 |
+
"eval_loss": 0.576337456703186,
|
| 870 |
+
"eval_runtime": 138.4638,
|
| 871 |
+
"eval_samples_per_second": 128.279,
|
| 872 |
+
"eval_steps_per_second": 4.015,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 8.626097867001254,
|
| 877 |
+
"grad_norm": 1.9795759916305542,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.5783,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 8.626097867001254,
|
| 884 |
+
"eval_accuracy": 0.8727814942897176,
|
| 885 |
+
"eval_loss": 0.5744160413742065,
|
| 886 |
+
"eval_runtime": 138.6602,
|
| 887 |
+
"eval_samples_per_second": 128.097,
|
| 888 |
+
"eval_steps_per_second": 4.01,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 8.782936010037641,
|
| 893 |
+
"grad_norm": 2.045375108718872,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.5722,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 8.782936010037641,
|
| 900 |
+
"eval_accuracy": 0.8730765855434479,
|
| 901 |
+
"eval_loss": 0.5763044357299805,
|
| 902 |
+
"eval_runtime": 138.2007,
|
| 903 |
+
"eval_samples_per_second": 128.523,
|
| 904 |
+
"eval_steps_per_second": 4.023,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 8.939774153074028,
|
| 909 |
+
"grad_norm": 2.2058000564575195,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.5757,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 8.939774153074028,
|
| 916 |
+
"eval_accuracy": 0.8740224639158709,
|
| 917 |
+
"eval_loss": 0.5730984807014465,
|
| 918 |
+
"eval_runtime": 136.766,
|
| 919 |
+
"eval_samples_per_second": 129.871,
|
| 920 |
+
"eval_steps_per_second": 4.065,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 9.096612296110415,
|
| 925 |
+
"grad_norm": 1.7985414266586304,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.57,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 9.096612296110415,
|
| 932 |
+
"eval_accuracy": 0.8736375865236091,
|
| 933 |
+
"eval_loss": 0.5691736340522766,
|
| 934 |
+
"eval_runtime": 138.8542,
|
| 935 |
+
"eval_samples_per_second": 127.918,
|
| 936 |
+
"eval_steps_per_second": 4.004,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 9.2534504391468,
|
| 941 |
+
"grad_norm": 1.985169529914856,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.5676,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 9.2534504391468,
|
| 948 |
+
"eval_accuracy": 0.8750337101060947,
|
| 949 |
+
"eval_loss": 0.5675247311592102,
|
| 950 |
+
"eval_runtime": 136.9465,
|
| 951 |
+
"eval_samples_per_second": 129.7,
|
| 952 |
+
"eval_steps_per_second": 4.06,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 9.410288582183187,
|
| 957 |
+
"grad_norm": 1.9350680112838745,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.5623,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 9.410288582183187,
|
| 964 |
+
"eval_accuracy": 0.8751417385906937,
|
| 965 |
+
"eval_loss": 0.565719723701477,
|
| 966 |
+
"eval_runtime": 137.9286,
|
| 967 |
+
"eval_samples_per_second": 128.777,
|
| 968 |
+
"eval_steps_per_second": 4.031,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 9.567126725219573,
|
| 973 |
+
"grad_norm": 2.2636570930480957,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.5607,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 9.567126725219573,
|
| 980 |
+
"eval_accuracy": 0.8758738273486182,
|
| 981 |
+
"eval_loss": 0.5632455945014954,
|
| 982 |
+
"eval_runtime": 137.8656,
|
| 983 |
+
"eval_samples_per_second": 128.836,
|
| 984 |
+
"eval_steps_per_second": 4.033,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 9.72396486825596,
|
| 989 |
+
"grad_norm": 2.0560171604156494,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.5624,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 9.72396486825596,
|
| 996 |
+
"eval_accuracy": 0.8763654574206416,
|
| 997 |
+
"eval_loss": 0.5634691119194031,
|
| 998 |
+
"eval_runtime": 139.2611,
|
| 999 |
+
"eval_samples_per_second": 127.545,
|
| 1000 |
+
"eval_steps_per_second": 3.993,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 9.880803011292347,
|
| 1005 |
+
"grad_norm": 2.1383323669433594,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.5578,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 9.880803011292347,
|
| 1012 |
+
"eval_accuracy": 0.8765358782549506,
|
| 1013 |
+
"eval_loss": 0.5544874668121338,
|
| 1014 |
+
"eval_runtime": 135.302,
|
| 1015 |
+
"eval_samples_per_second": 131.277,
|
| 1016 |
+
"eval_steps_per_second": 4.109,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 10.037641154328734,
|
| 1021 |
+
"grad_norm": 1.7943627834320068,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.556,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 10.037641154328734,
|
| 1028 |
+
"eval_accuracy": 0.8763150725578216,
|
| 1029 |
+
"eval_loss": 0.5644202828407288,
|
| 1030 |
+
"eval_runtime": 136.08,
|
| 1031 |
+
"eval_samples_per_second": 130.526,
|
| 1032 |
+
"eval_steps_per_second": 4.086,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 10.194479297365119,
|
| 1037 |
+
"grad_norm": 1.7182546854019165,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.5535,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 10.194479297365119,
|
| 1044 |
+
"eval_accuracy": 0.8764190942778233,
|
| 1045 |
+
"eval_loss": 0.5606905817985535,
|
| 1046 |
+
"eval_runtime": 135.1049,
|
| 1047 |
+
"eval_samples_per_second": 131.468,
|
| 1048 |
+
"eval_steps_per_second": 4.115,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 10.351317440401505,
|
| 1053 |
+
"grad_norm": 1.8069394826889038,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.5522,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 10.351317440401505,
|
| 1060 |
+
"eval_accuracy": 0.8765611690269617,
|
| 1061 |
+
"eval_loss": 0.5583159327507019,
|
| 1062 |
+
"eval_runtime": 135.3996,
|
| 1063 |
+
"eval_samples_per_second": 131.182,
|
| 1064 |
+
"eval_steps_per_second": 4.106,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 10.508155583437892,
|
| 1069 |
+
"grad_norm": 4.351428985595703,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.5478,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 10.508155583437892,
|
| 1076 |
+
"eval_accuracy": 0.8769738063470637,
|
| 1077 |
+
"eval_loss": 0.5574548244476318,
|
| 1078 |
+
"eval_runtime": 142.6768,
|
| 1079 |
+
"eval_samples_per_second": 124.491,
|
| 1080 |
+
"eval_steps_per_second": 3.897,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 10.664993726474279,
|
| 1085 |
+
"grad_norm": 2.1382317543029785,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.5495,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 10.664993726474279,
|
| 1092 |
+
"eval_accuracy": 0.87792866101614,
|
| 1093 |
+
"eval_loss": 0.5549121499061584,
|
| 1094 |
+
"eval_runtime": 141.5749,
|
| 1095 |
+
"eval_samples_per_second": 125.46,
|
| 1096 |
+
"eval_steps_per_second": 3.927,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 10.821831869510666,
|
| 1101 |
+
"grad_norm": 1.9487907886505127,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.5461,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 10.821831869510666,
|
| 1108 |
+
"eval_accuracy": 0.8785157944038718,
|
| 1109 |
+
"eval_loss": 0.5503472089767456,
|
| 1110 |
+
"eval_runtime": 137.67,
|
| 1111 |
+
"eval_samples_per_second": 129.019,
|
| 1112 |
+
"eval_steps_per_second": 4.039,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 10.97867001254705,
|
| 1117 |
+
"grad_norm": 1.764410376548767,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.5423,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 10.97867001254705,
|
| 1124 |
+
"eval_accuracy": 0.8787178107864233,
|
| 1125 |
+
"eval_loss": 0.5458616018295288,
|
| 1126 |
+
"eval_runtime": 136.6703,
|
| 1127 |
+
"eval_samples_per_second": 129.962,
|
| 1128 |
+
"eval_steps_per_second": 4.068,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 11.135508155583437,
|
| 1133 |
+
"grad_norm": 1.784921646118164,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.538,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 11.135508155583437,
|
| 1140 |
+
"eval_accuracy": 0.8794234085971385,
|
| 1141 |
+
"eval_loss": 0.545362651348114,
|
| 1142 |
+
"eval_runtime": 136.2873,
|
| 1143 |
+
"eval_samples_per_second": 130.328,
|
| 1144 |
+
"eval_steps_per_second": 4.08,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 11.292346298619824,
|
| 1149 |
+
"grad_norm": 1.745466947555542,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.534,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 11.292346298619824,
|
| 1156 |
+
"eval_accuracy": 0.8792399378910559,
|
| 1157 |
+
"eval_loss": 0.5446943640708923,
|
| 1158 |
+
"eval_runtime": 136.111,
|
| 1159 |
+
"eval_samples_per_second": 130.496,
|
| 1160 |
+
"eval_steps_per_second": 4.085,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 11.449184441656211,
|
| 1165 |
+
"grad_norm": 1.9506378173828125,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.5344,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 11.449184441656211,
|
| 1172 |
+
"eval_accuracy": 0.8796105263431989,
|
| 1173 |
+
"eval_loss": 0.5461797118186951,
|
| 1174 |
+
"eval_runtime": 135.9492,
|
| 1175 |
+
"eval_samples_per_second": 130.652,
|
| 1176 |
+
"eval_steps_per_second": 4.09,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 11.606022584692598,
|
| 1181 |
+
"grad_norm": 1.815746784210205,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.5299,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 11.606022584692598,
|
| 1188 |
+
"eval_accuracy": 0.8803022028035682,
|
| 1189 |
+
"eval_loss": 0.5455493927001953,
|
| 1190 |
+
"eval_runtime": 138.5251,
|
| 1191 |
+
"eval_samples_per_second": 128.222,
|
| 1192 |
+
"eval_steps_per_second": 4.014,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 11.762860727728985,
|
| 1197 |
+
"grad_norm": 1.744718313217163,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.5396,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 11.762860727728985,
|
| 1204 |
+
"eval_accuracy": 0.8805864915226863,
|
| 1205 |
+
"eval_loss": 0.5399504899978638,
|
| 1206 |
+
"eval_runtime": 137.1238,
|
| 1207 |
+
"eval_samples_per_second": 129.533,
|
| 1208 |
+
"eval_steps_per_second": 4.055,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 11.91969887076537,
|
| 1213 |
+
"grad_norm": 2.1070852279663086,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.5382,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 11.91969887076537,
|
| 1220 |
+
"eval_accuracy": 0.8815948803048729,
|
| 1221 |
+
"eval_loss": 0.5347013473510742,
|
| 1222 |
+
"eval_runtime": 135.8625,
|
| 1223 |
+
"eval_samples_per_second": 130.735,
|
| 1224 |
+
"eval_steps_per_second": 4.092,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 12.076537013801756,
|
| 1229 |
+
"grad_norm": 1.5224947929382324,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.5324,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 12.076537013801756,
|
| 1236 |
+
"eval_accuracy": 0.8813583003628418,
|
| 1237 |
+
"eval_loss": 0.5384557843208313,
|
| 1238 |
+
"eval_runtime": 135.7603,
|
| 1239 |
+
"eval_samples_per_second": 130.833,
|
| 1240 |
+
"eval_steps_per_second": 4.095,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 12.233375156838143,
|
| 1245 |
+
"grad_norm": 1.964902400970459,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.5226,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 12.233375156838143,
|
| 1252 |
+
"eval_accuracy": 0.8812442762338378,
|
| 1253 |
+
"eval_loss": 0.5403480529785156,
|
| 1254 |
+
"eval_runtime": 137.0414,
|
| 1255 |
+
"eval_samples_per_second": 129.61,
|
| 1256 |
+
"eval_steps_per_second": 4.057,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 12.39021329987453,
|
| 1261 |
+
"grad_norm": 1.7547788619995117,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.5309,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 12.39021329987453,
|
| 1268 |
+
"eval_accuracy": 0.8812615536828584,
|
| 1269 |
+
"eval_loss": 0.5354346036911011,
|
| 1270 |
+
"eval_runtime": 135.9832,
|
| 1271 |
+
"eval_samples_per_second": 130.619,
|
| 1272 |
+
"eval_steps_per_second": 4.089,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 12.547051442910917,
|
| 1277 |
+
"grad_norm": 1.6900429725646973,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.5226,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 12.547051442910917,
|
| 1284 |
+
"eval_accuracy": 0.8825562243792318,
|
| 1285 |
+
"eval_loss": 0.5320644378662109,
|
| 1286 |
+
"eval_runtime": 139.0622,
|
| 1287 |
+
"eval_samples_per_second": 127.727,
|
| 1288 |
+
"eval_steps_per_second": 3.998,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 12.703889585947302,
|
| 1293 |
+
"grad_norm": 1.8609619140625,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.5202,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 12.703889585947302,
|
| 1300 |
+
"eval_accuracy": 0.8831901242109903,
|
| 1301 |
+
"eval_loss": 0.5275829434394836,
|
| 1302 |
+
"eval_runtime": 136.289,
|
| 1303 |
+
"eval_samples_per_second": 130.326,
|
| 1304 |
+
"eval_steps_per_second": 4.08,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 12.860727728983688,
|
| 1309 |
+
"grad_norm": 1.7507585287094116,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.5226,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 12.860727728983688,
|
| 1316 |
+
"eval_accuracy": 0.8828808954726105,
|
| 1317 |
+
"eval_loss": 0.5298667550086975,
|
| 1318 |
+
"eval_runtime": 136.2416,
|
| 1319 |
+
"eval_samples_per_second": 130.371,
|
| 1320 |
+
"eval_steps_per_second": 4.081,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 13.017565872020075,
|
| 1325 |
+
"grad_norm": 1.6217900514602661,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.5195,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 13.017565872020075,
|
| 1332 |
+
"eval_accuracy": 0.8829326864682345,
|
| 1333 |
+
"eval_loss": 0.5272156596183777,
|
| 1334 |
+
"eval_runtime": 136.4776,
|
| 1335 |
+
"eval_samples_per_second": 130.146,
|
| 1336 |
+
"eval_steps_per_second": 4.074,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 13.174404015056462,
|
| 1341 |
+
"grad_norm": 1.7102445363998413,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.5156,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 13.174404015056462,
|
| 1348 |
+
"eval_accuracy": 0.8827681324734242,
|
| 1349 |
+
"eval_loss": 0.5329484343528748,
|
| 1350 |
+
"eval_runtime": 136.9515,
|
| 1351 |
+
"eval_samples_per_second": 129.696,
|
| 1352 |
+
"eval_steps_per_second": 4.06,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 13.331242158092849,
|
| 1357 |
+
"grad_norm": 1.986724853515625,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.5089,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 13.331242158092849,
|
| 1364 |
+
"eval_accuracy": 0.883174400510713,
|
| 1365 |
+
"eval_loss": 0.53280109167099,
|
| 1366 |
+
"eval_runtime": 137.7862,
|
| 1367 |
+
"eval_samples_per_second": 128.91,
|
| 1368 |
+
"eval_steps_per_second": 4.035,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 13.488080301129235,
|
| 1373 |
+
"grad_norm": 1.8194265365600586,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.5116,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 13.488080301129235,
|
| 1380 |
+
"eval_accuracy": 0.8834585911212417,
|
| 1381 |
+
"eval_loss": 0.5313114523887634,
|
| 1382 |
+
"eval_runtime": 135.4903,
|
| 1383 |
+
"eval_samples_per_second": 131.094,
|
| 1384 |
+
"eval_steps_per_second": 4.104,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 13.64491844416562,
|
| 1389 |
+
"grad_norm": 1.627774715423584,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.5111,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 13.64491844416562,
|
| 1396 |
+
"eval_accuracy": 0.8836516217774597,
|
| 1397 |
+
"eval_loss": 0.5313425064086914,
|
| 1398 |
+
"eval_runtime": 135.5417,
|
| 1399 |
+
"eval_samples_per_second": 131.044,
|
| 1400 |
+
"eval_steps_per_second": 4.102,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 13.801756587202007,
|
| 1405 |
+
"grad_norm": 1.8606806993484497,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.514,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 13.801756587202007,
|
| 1412 |
+
"eval_accuracy": 0.8845101168592323,
|
| 1413 |
+
"eval_loss": 0.5231484770774841,
|
| 1414 |
+
"eval_runtime": 135.6215,
|
| 1415 |
+
"eval_samples_per_second": 130.967,
|
| 1416 |
+
"eval_steps_per_second": 4.1,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 13.958594730238394,
|
| 1421 |
+
"grad_norm": 1.6027812957763672,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.5126,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 13.958594730238394,
|
| 1428 |
+
"eval_accuracy": 0.8856740479715313,
|
| 1429 |
+
"eval_loss": 0.519383430480957,
|
| 1430 |
+
"eval_runtime": 135.5201,
|
| 1431 |
+
"eval_samples_per_second": 131.065,
|
| 1432 |
+
"eval_steps_per_second": 4.103,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 14.11543287327478,
|
| 1437 |
+
"grad_norm": 1.834935188293457,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.5077,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 14.11543287327478,
|
| 1444 |
+
"eval_accuracy": 0.8853411508309706,
|
| 1445 |
+
"eval_loss": 0.5175889730453491,
|
| 1446 |
+
"eval_runtime": 136.628,
|
| 1447 |
+
"eval_samples_per_second": 130.003,
|
| 1448 |
+
"eval_steps_per_second": 4.069,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 14.272271016311167,
|
| 1453 |
+
"grad_norm": 1.7725075483322144,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.5047,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 14.272271016311167,
|
| 1460 |
+
"eval_accuracy": 0.885856389775109,
|
| 1461 |
+
"eval_loss": 0.518938422203064,
|
| 1462 |
+
"eval_runtime": 135.6215,
|
| 1463 |
+
"eval_samples_per_second": 130.967,
|
| 1464 |
+
"eval_steps_per_second": 4.1,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 14.429109159347554,
|
| 1469 |
+
"grad_norm": 1.8045620918273926,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.5006,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 14.429109159347554,
|
| 1476 |
+
"eval_accuracy": 0.8857257240219224,
|
| 1477 |
+
"eval_loss": 0.5159721970558167,
|
| 1478 |
+
"eval_runtime": 136.7371,
|
| 1479 |
+
"eval_samples_per_second": 129.899,
|
| 1480 |
+
"eval_steps_per_second": 4.066,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 14.58594730238394,
|
| 1485 |
+
"grad_norm": 1.666648268699646,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.5006,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 14.58594730238394,
|
| 1492 |
+
"eval_accuracy": 0.8859774237990279,
|
| 1493 |
+
"eval_loss": 0.5151115655899048,
|
| 1494 |
+
"eval_runtime": 135.4637,
|
| 1495 |
+
"eval_samples_per_second": 131.12,
|
| 1496 |
+
"eval_steps_per_second": 4.104,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 14.742785445420326,
|
| 1501 |
+
"grad_norm": 2.182983875274658,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.5001,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 14.742785445420326,
|
| 1508 |
+
"eval_accuracy": 0.8863202870681423,
|
| 1509 |
+
"eval_loss": 0.5123385787010193,
|
| 1510 |
+
"eval_runtime": 135.8256,
|
| 1511 |
+
"eval_samples_per_second": 130.771,
|
| 1512 |
+
"eval_steps_per_second": 4.093,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 14.899623588456713,
|
| 1517 |
+
"grad_norm": 1.8055123090744019,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.4997,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 14.899623588456713,
|
| 1524 |
+
"eval_accuracy": 0.8863326569897508,
|
| 1525 |
+
"eval_loss": 0.5162584781646729,
|
| 1526 |
+
"eval_runtime": 136.6447,
|
| 1527 |
+
"eval_samples_per_second": 129.987,
|
| 1528 |
+
"eval_steps_per_second": 4.069,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 15.0564617314931,
|
| 1533 |
+
"grad_norm": 1.76173734664917,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.4944,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 15.0564617314931,
|
| 1540 |
+
"eval_accuracy": 0.8870298067922737,
|
| 1541 |
+
"eval_loss": 0.5153139233589172,
|
| 1542 |
+
"eval_runtime": 135.4914,
|
| 1543 |
+
"eval_samples_per_second": 131.093,
|
| 1544 |
+
"eval_steps_per_second": 4.104,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 15.213299874529486,
|
| 1549 |
+
"grad_norm": 1.8028675317764282,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.4931,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 15.213299874529486,
|
| 1556 |
+
"eval_accuracy": 0.8870491203172791,
|
| 1557 |
+
"eval_loss": 0.5170115828514099,
|
| 1558 |
+
"eval_runtime": 135.504,
|
| 1559 |
+
"eval_samples_per_second": 131.081,
|
| 1560 |
+
"eval_steps_per_second": 4.103,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 15.370138017565871,
|
| 1565 |
+
"grad_norm": 1.8522734642028809,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.4895,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 15.370138017565871,
|
| 1572 |
+
"eval_accuracy": 0.8872314989109974,
|
| 1573 |
+
"eval_loss": 0.5082823634147644,
|
| 1574 |
+
"eval_runtime": 135.6928,
|
| 1575 |
+
"eval_samples_per_second": 130.899,
|
| 1576 |
+
"eval_steps_per_second": 4.097,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 15.526976160602258,
|
| 1581 |
+
"grad_norm": 2.180356502532959,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.4939,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 15.526976160602258,
|
| 1588 |
+
"eval_accuracy": 0.8873179448478067,
|
| 1589 |
+
"eval_loss": 0.5147595405578613,
|
| 1590 |
+
"eval_runtime": 136.8272,
|
| 1591 |
+
"eval_samples_per_second": 129.813,
|
| 1592 |
+
"eval_steps_per_second": 4.064,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 15.683814303638645,
|
| 1597 |
+
"grad_norm": 1.905341625213623,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.4901,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 15.683814303638645,
|
| 1604 |
+
"eval_accuracy": 0.8888588658773607,
|
| 1605 |
+
"eval_loss": 0.5002791285514832,
|
| 1606 |
+
"eval_runtime": 138.1977,
|
| 1607 |
+
"eval_samples_per_second": 128.526,
|
| 1608 |
+
"eval_steps_per_second": 4.023,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 15.840652446675032,
|
| 1613 |
+
"grad_norm": 1.8882616758346558,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.4872,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 15.840652446675032,
|
| 1620 |
+
"eval_accuracy": 0.8882201726503743,
|
| 1621 |
+
"eval_loss": 0.5107588171958923,
|
| 1622 |
+
"eval_runtime": 137.2768,
|
| 1623 |
+
"eval_samples_per_second": 129.388,
|
| 1624 |
+
"eval_steps_per_second": 4.05,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 15.997490589711418,
|
| 1629 |
+
"grad_norm": 1.8350785970687866,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.4886,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 15.997490589711418,
|
| 1636 |
+
"eval_accuracy": 0.8884791331148998,
|
| 1637 |
+
"eval_loss": 0.5093386769294739,
|
| 1638 |
+
"eval_runtime": 138.8052,
|
| 1639 |
+
"eval_samples_per_second": 127.963,
|
| 1640 |
+
"eval_steps_per_second": 4.006,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 16.154328732747803,
|
| 1645 |
+
"grad_norm": 1.8488084077835083,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.4851,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 16.154328732747803,
|
| 1652 |
+
"eval_accuracy": 0.8891495748698329,
|
| 1653 |
+
"eval_loss": 0.5062207579612732,
|
| 1654 |
+
"eval_runtime": 135.8929,
|
| 1655 |
+
"eval_samples_per_second": 130.706,
|
| 1656 |
+
"eval_steps_per_second": 4.091,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 16.311166875784192,
|
| 1661 |
+
"grad_norm": 1.8075292110443115,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.4798,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 16.311166875784192,
|
| 1668 |
+
"eval_accuracy": 0.8890000217056283,
|
| 1669 |
+
"eval_loss": 0.5002816319465637,
|
| 1670 |
+
"eval_runtime": 137.0292,
|
| 1671 |
+
"eval_samples_per_second": 129.622,
|
| 1672 |
+
"eval_steps_per_second": 4.058,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 16.468005018820577,
|
| 1677 |
+
"grad_norm": 2.047656297683716,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.4817,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 16.468005018820577,
|
| 1684 |
+
"eval_accuracy": 0.8886614267485646,
|
| 1685 |
+
"eval_loss": 0.5042737126350403,
|
| 1686 |
+
"eval_runtime": 136.0602,
|
| 1687 |
+
"eval_samples_per_second": 130.545,
|
| 1688 |
+
"eval_steps_per_second": 4.086,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 16.624843161856965,
|
| 1693 |
+
"grad_norm": 1.7578613758087158,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.4768,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 16.624843161856965,
|
| 1700 |
+
"eval_accuracy": 0.8893812530489446,
|
| 1701 |
+
"eval_loss": 0.502278208732605,
|
| 1702 |
+
"eval_runtime": 136.2702,
|
| 1703 |
+
"eval_samples_per_second": 130.344,
|
| 1704 |
+
"eval_steps_per_second": 4.08,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 16.78168130489335,
|
| 1709 |
+
"grad_norm": 2.1175436973571777,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.4828,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 16.78168130489335,
|
| 1716 |
+
"eval_accuracy": 0.8899669997369356,
|
| 1717 |
+
"eval_loss": 0.4929199516773224,
|
| 1718 |
+
"eval_runtime": 136.0653,
|
| 1719 |
+
"eval_samples_per_second": 130.54,
|
| 1720 |
+
"eval_steps_per_second": 4.086,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 16.938519447929735,
|
| 1725 |
+
"grad_norm": 1.8378137350082397,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.484,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 16.938519447929735,
|
| 1732 |
+
"eval_accuracy": 0.8901911152902642,
|
| 1733 |
+
"eval_loss": 0.4961177110671997,
|
| 1734 |
+
"eval_runtime": 136.872,
|
| 1735 |
+
"eval_samples_per_second": 129.771,
|
| 1736 |
+
"eval_steps_per_second": 4.062,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 17.095357590966124,
|
| 1741 |
+
"grad_norm": 1.5262337923049927,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.4751,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 17.095357590966124,
|
| 1748 |
+
"eval_accuracy": 0.8903900552457398,
|
| 1749 |
+
"eval_loss": 0.49581587314605713,
|
| 1750 |
+
"eval_runtime": 135.9798,
|
| 1751 |
+
"eval_samples_per_second": 130.622,
|
| 1752 |
+
"eval_steps_per_second": 4.089,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 17.25219573400251,
|
| 1757 |
+
"grad_norm": 1.6708686351776123,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.4725,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 17.25219573400251,
|
| 1764 |
+
"eval_accuracy": 0.8904223785784117,
|
| 1765 |
+
"eval_loss": 0.49634990096092224,
|
| 1766 |
+
"eval_runtime": 135.8739,
|
| 1767 |
+
"eval_samples_per_second": 130.724,
|
| 1768 |
+
"eval_steps_per_second": 4.092,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 17.409033877038897,
|
| 1773 |
+
"grad_norm": 1.7331116199493408,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.471,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 17.409033877038897,
|
| 1780 |
+
"eval_accuracy": 0.8903350047991724,
|
| 1781 |
+
"eval_loss": 0.4966423511505127,
|
| 1782 |
+
"eval_runtime": 135.8945,
|
| 1783 |
+
"eval_samples_per_second": 130.704,
|
| 1784 |
+
"eval_steps_per_second": 4.091,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 17.565872020075282,
|
| 1789 |
+
"grad_norm": 2.2375760078430176,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.4719,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 17.565872020075282,
|
| 1796 |
+
"eval_accuracy": 0.8906306370526978,
|
| 1797 |
+
"eval_loss": 0.49662166833877563,
|
| 1798 |
+
"eval_runtime": 137.6637,
|
| 1799 |
+
"eval_samples_per_second": 129.025,
|
| 1800 |
+
"eval_steps_per_second": 4.039,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 17.722710163111667,
|
| 1805 |
+
"grad_norm": 1.9899126291275024,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.4726,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 17.722710163111667,
|
| 1812 |
+
"eval_accuracy": 0.891625632887741,
|
| 1813 |
+
"eval_loss": 0.4905398488044739,
|
| 1814 |
+
"eval_runtime": 135.7586,
|
| 1815 |
+
"eval_samples_per_second": 130.835,
|
| 1816 |
+
"eval_steps_per_second": 4.096,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 17.879548306148056,
|
| 1821 |
+
"grad_norm": 1.6155794858932495,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.4703,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 17.879548306148056,
|
| 1828 |
+
"eval_accuracy": 0.8915143558898375,
|
| 1829 |
+
"eval_loss": 0.4848301410675049,
|
| 1830 |
+
"eval_runtime": 135.9168,
|
| 1831 |
+
"eval_samples_per_second": 130.683,
|
| 1832 |
+
"eval_steps_per_second": 4.091,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 18.03638644918444,
|
| 1837 |
+
"grad_norm": 1.7314578294754028,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.4681,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 18.03638644918444,
|
| 1844 |
+
"eval_accuracy": 0.8920219108886556,
|
| 1845 |
+
"eval_loss": 0.4843462407588959,
|
| 1846 |
+
"eval_runtime": 135.7068,
|
| 1847 |
+
"eval_samples_per_second": 130.885,
|
| 1848 |
+
"eval_steps_per_second": 4.097,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 18.19322459222083,
|
| 1853 |
+
"grad_norm": 1.4559590816497803,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.4653,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 18.19322459222083,
|
| 1860 |
+
"eval_accuracy": 0.8917177347534987,
|
| 1861 |
+
"eval_loss": 0.48741909861564636,
|
| 1862 |
+
"eval_runtime": 136.4404,
|
| 1863 |
+
"eval_samples_per_second": 130.181,
|
| 1864 |
+
"eval_steps_per_second": 4.075,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 18.350062735257215,
|
| 1869 |
+
"grad_norm": 1.9733667373657227,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.467,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 18.350062735257215,
|
| 1876 |
+
"eval_accuracy": 0.8921875868103784,
|
| 1877 |
+
"eval_loss": 0.4918387532234192,
|
| 1878 |
+
"eval_runtime": 137.5074,
|
| 1879 |
+
"eval_samples_per_second": 129.171,
|
| 1880 |
+
"eval_steps_per_second": 4.043,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 18.5069008782936,
|
| 1885 |
+
"grad_norm": 1.848181962966919,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.4614,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 18.5069008782936,
|
| 1892 |
+
"eval_accuracy": 0.892308105574104,
|
| 1893 |
+
"eval_loss": 0.48882216215133667,
|
| 1894 |
+
"eval_runtime": 137.8932,
|
| 1895 |
+
"eval_samples_per_second": 128.81,
|
| 1896 |
+
"eval_steps_per_second": 4.032,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 18.663739021329988,
|
| 1901 |
+
"grad_norm": 1.4978140592575073,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.4622,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 18.663739021329988,
|
| 1908 |
+
"eval_accuracy": 0.8926432062527199,
|
| 1909 |
+
"eval_loss": 0.48588448762893677,
|
| 1910 |
+
"eval_runtime": 138.2065,
|
| 1911 |
+
"eval_samples_per_second": 128.518,
|
| 1912 |
+
"eval_steps_per_second": 4.023,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 18.820577164366373,
|
| 1917 |
+
"grad_norm": 1.6420320272445679,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.4639,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 18.820577164366373,
|
| 1924 |
+
"eval_accuracy": 0.8927606591282531,
|
| 1925 |
+
"eval_loss": 0.4796488285064697,
|
| 1926 |
+
"eval_runtime": 138.0236,
|
| 1927 |
+
"eval_samples_per_second": 128.688,
|
| 1928 |
+
"eval_steps_per_second": 4.028,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 18.97741530740276,
|
| 1933 |
+
"grad_norm": 1.7469255924224854,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.4658,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 18.97741530740276,
|
| 1940 |
+
"eval_accuracy": 0.8922558893384064,
|
| 1941 |
+
"eval_loss": 0.48980244994163513,
|
| 1942 |
+
"eval_runtime": 139.102,
|
| 1943 |
+
"eval_samples_per_second": 127.691,
|
| 1944 |
+
"eval_steps_per_second": 3.997,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 19.134253450439147,
|
| 1949 |
+
"grad_norm": 1.4715181589126587,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.4575,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 19.134253450439147,
|
| 1956 |
+
"eval_accuracy": 0.8931160159325203,
|
| 1957 |
+
"eval_loss": 0.487956702709198,
|
| 1958 |
+
"eval_runtime": 136.6881,
|
| 1959 |
+
"eval_samples_per_second": 129.946,
|
| 1960 |
+
"eval_steps_per_second": 4.068,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 19.29109159347553,
|
| 1965 |
+
"grad_norm": 1.3208478689193726,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.4552,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 19.29109159347553,
|
| 1972 |
+
"eval_accuracy": 0.8935916361299734,
|
| 1973 |
+
"eval_loss": 0.4730677306652069,
|
| 1974 |
+
"eval_runtime": 138.2656,
|
| 1975 |
+
"eval_samples_per_second": 128.463,
|
| 1976 |
+
"eval_steps_per_second": 4.021,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 19.44792973651192,
|
| 1981 |
+
"grad_norm": 1.6651636362075806,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.4578,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 19.44792973651192,
|
| 1988 |
+
"eval_accuracy": 0.8929142807565504,
|
| 1989 |
+
"eval_loss": 0.4812925159931183,
|
| 1990 |
+
"eval_runtime": 136.12,
|
| 1991 |
+
"eval_samples_per_second": 130.488,
|
| 1992 |
+
"eval_steps_per_second": 4.085,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 19.604767879548305,
|
| 1997 |
+
"grad_norm": 1.7527645826339722,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.4608,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 19.604767879548305,
|
| 2004 |
+
"eval_accuracy": 0.8937893734698642,
|
| 2005 |
+
"eval_loss": 0.48109447956085205,
|
| 2006 |
+
"eval_runtime": 135.5275,
|
| 2007 |
+
"eval_samples_per_second": 131.058,
|
| 2008 |
+
"eval_steps_per_second": 4.102,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 19.761606022584694,
|
| 2013 |
+
"grad_norm": 1.9757037162780762,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.4537,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 19.761606022584694,
|
| 2020 |
+
"eval_accuracy": 0.8938628190167149,
|
| 2021 |
+
"eval_loss": 0.48102205991744995,
|
| 2022 |
+
"eval_runtime": 135.9358,
|
| 2023 |
+
"eval_samples_per_second": 130.665,
|
| 2024 |
+
"eval_steps_per_second": 4.09,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 19.91844416562108,
|
| 2029 |
+
"grad_norm": 2.1024065017700195,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.4504,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 19.91844416562108,
|
| 2036 |
+
"eval_accuracy": 0.8941597309290022,
|
| 2037 |
+
"eval_loss": 0.48300912976264954,
|
| 2038 |
+
"eval_runtime": 136.0232,
|
| 2039 |
+
"eval_samples_per_second": 130.581,
|
| 2040 |
+
"eval_steps_per_second": 4.088,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 20.075282308657467,
|
| 2045 |
+
"grad_norm": 1.5361249446868896,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.4521,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 20.075282308657467,
|
| 2052 |
+
"eval_accuracy": 0.8941135671473668,
|
| 2053 |
+
"eval_loss": 0.4824671745300293,
|
| 2054 |
+
"eval_runtime": 136.7461,
|
| 2055 |
+
"eval_samples_per_second": 129.89,
|
| 2056 |
+
"eval_steps_per_second": 4.066,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 20.232120451693852,
|
| 2061 |
+
"grad_norm": 1.7190461158752441,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.4482,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 20.232120451693852,
|
| 2068 |
+
"eval_accuracy": 0.8942540949009935,
|
| 2069 |
+
"eval_loss": 0.4761606454849243,
|
| 2070 |
+
"eval_runtime": 135.8116,
|
| 2071 |
+
"eval_samples_per_second": 130.784,
|
| 2072 |
+
"eval_steps_per_second": 4.094,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 20.388958594730237,
|
| 2077 |
+
"grad_norm": 1.773404598236084,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.452,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 20.388958594730237,
|
| 2084 |
+
"eval_accuracy": 0.8940873264670252,
|
| 2085 |
+
"eval_loss": 0.4781154692173004,
|
| 2086 |
+
"eval_runtime": 137.0134,
|
| 2087 |
+
"eval_samples_per_second": 129.637,
|
| 2088 |
+
"eval_steps_per_second": 4.058,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 20.545796737766626,
|
| 2093 |
+
"grad_norm": 1.8081912994384766,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.4484,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 20.545796737766626,
|
| 2100 |
+
"eval_accuracy": 0.8946489572135398,
|
| 2101 |
+
"eval_loss": 0.4731599688529968,
|
| 2102 |
+
"eval_runtime": 135.6427,
|
| 2103 |
+
"eval_samples_per_second": 130.947,
|
| 2104 |
+
"eval_steps_per_second": 4.099,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 20.70263488080301,
|
| 2109 |
+
"grad_norm": 1.6236828565597534,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.4447,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 20.70263488080301,
|
| 2116 |
+
"eval_accuracy": 0.8951635225528451,
|
| 2117 |
+
"eval_loss": 0.475312203168869,
|
| 2118 |
+
"eval_runtime": 136.7392,
|
| 2119 |
+
"eval_samples_per_second": 129.897,
|
| 2120 |
+
"eval_steps_per_second": 4.066,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 20.8594730238394,
|
| 2125 |
+
"grad_norm": 1.79032301902771,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.4475,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 20.8594730238394,
|
| 2132 |
+
"eval_accuracy": 0.8947900526569977,
|
| 2133 |
+
"eval_loss": 0.4722547233104706,
|
| 2134 |
+
"eval_runtime": 135.9539,
|
| 2135 |
+
"eval_samples_per_second": 130.647,
|
| 2136 |
+
"eval_steps_per_second": 4.09,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 21.016311166875784,
|
| 2141 |
+
"grad_norm": 1.848694920539856,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.4453,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 21.016311166875784,
|
| 2148 |
+
"eval_accuracy": 0.8956351519188471,
|
| 2149 |
+
"eval_loss": 0.47477635741233826,
|
| 2150 |
+
"eval_runtime": 136.9081,
|
| 2151 |
+
"eval_samples_per_second": 129.737,
|
| 2152 |
+
"eval_steps_per_second": 4.061,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 21.17314930991217,
|
| 2157 |
+
"grad_norm": 1.6243959665298462,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.4404,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 21.17314930991217,
|
| 2164 |
+
"eval_accuracy": 0.8954576844063186,
|
| 2165 |
+
"eval_loss": 0.47416919469833374,
|
| 2166 |
+
"eval_runtime": 135.9106,
|
| 2167 |
+
"eval_samples_per_second": 130.689,
|
| 2168 |
+
"eval_steps_per_second": 4.091,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 21.329987452948558,
|
| 2173 |
+
"grad_norm": 1.584000587463379,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.4447,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 21.329987452948558,
|
| 2180 |
+
"eval_accuracy": 0.8957169330654157,
|
| 2181 |
+
"eval_loss": 0.4676923453807831,
|
| 2182 |
+
"eval_runtime": 138.1913,
|
| 2183 |
+
"eval_samples_per_second": 128.532,
|
| 2184 |
+
"eval_steps_per_second": 4.023,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 21.486825595984943,
|
| 2189 |
+
"grad_norm": 1.5659821033477783,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.4432,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 21.486825595984943,
|
| 2196 |
+
"eval_accuracy": 0.8963371341718294,
|
| 2197 |
+
"eval_loss": 0.4697062075138092,
|
| 2198 |
+
"eval_runtime": 137.0564,
|
| 2199 |
+
"eval_samples_per_second": 129.596,
|
| 2200 |
+
"eval_steps_per_second": 4.057,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 21.64366373902133,
|
| 2205 |
+
"grad_norm": 1.9083049297332764,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.4347,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 21.64366373902133,
|
| 2212 |
+
"eval_accuracy": 0.8959190190096384,
|
| 2213 |
+
"eval_loss": 0.4709945619106293,
|
| 2214 |
+
"eval_runtime": 137.5493,
|
| 2215 |
+
"eval_samples_per_second": 129.132,
|
| 2216 |
+
"eval_steps_per_second": 4.042,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 21.800501882057716,
|
| 2221 |
+
"grad_norm": 2.0025572776794434,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.4409,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 21.800501882057716,
|
| 2228 |
+
"eval_accuracy": 0.8965652711228671,
|
| 2229 |
+
"eval_loss": 0.46794837713241577,
|
| 2230 |
+
"eval_runtime": 138.0531,
|
| 2231 |
+
"eval_samples_per_second": 128.661,
|
| 2232 |
+
"eval_steps_per_second": 4.027,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 21.9573400250941,
|
| 2237 |
+
"grad_norm": 1.7192397117614746,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.4369,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 21.9573400250941,
|
| 2244 |
+
"eval_accuracy": 0.8967704268678831,
|
| 2245 |
+
"eval_loss": 0.4641966223716736,
|
| 2246 |
+
"eval_runtime": 137.2779,
|
| 2247 |
+
"eval_samples_per_second": 129.387,
|
| 2248 |
+
"eval_steps_per_second": 4.05,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 22.11417816813049,
|
| 2253 |
+
"grad_norm": 2.3039047718048096,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.4324,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 22.11417816813049,
|
| 2260 |
+
"eval_accuracy": 0.8967945146663492,
|
| 2261 |
+
"eval_loss": 0.4679114520549774,
|
| 2262 |
+
"eval_runtime": 136.9237,
|
| 2263 |
+
"eval_samples_per_second": 129.722,
|
| 2264 |
+
"eval_steps_per_second": 4.061,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 22.271016311166875,
|
| 2269 |
+
"grad_norm": 1.5516401529312134,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.4391,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 22.271016311166875,
|
| 2276 |
+
"eval_accuracy": 0.8974568243601687,
|
| 2277 |
+
"eval_loss": 0.4623771607875824,
|
| 2278 |
+
"eval_runtime": 137.7406,
|
| 2279 |
+
"eval_samples_per_second": 128.953,
|
| 2280 |
+
"eval_steps_per_second": 4.037,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 22.427854454203263,
|
| 2285 |
+
"grad_norm": 1.6362214088439941,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.432,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 22.427854454203263,
|
| 2292 |
+
"eval_accuracy": 0.8975918781232312,
|
| 2293 |
+
"eval_loss": 0.4618777334690094,
|
| 2294 |
+
"eval_runtime": 136.3765,
|
| 2295 |
+
"eval_samples_per_second": 130.242,
|
| 2296 |
+
"eval_steps_per_second": 4.077,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 22.58469259723965,
|
| 2301 |
+
"grad_norm": 1.8618403673171997,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.4318,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 22.58469259723965,
|
| 2308 |
+
"eval_accuracy": 0.8970239624042927,
|
| 2309 |
+
"eval_loss": 0.4642651379108429,
|
| 2310 |
+
"eval_runtime": 135.7523,
|
| 2311 |
+
"eval_samples_per_second": 130.841,
|
| 2312 |
+
"eval_steps_per_second": 4.096,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 22.741530740276033,
|
| 2317 |
+
"grad_norm": 2.17258620262146,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.4301,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 22.741530740276033,
|
| 2324 |
+
"eval_accuracy": 0.8972558582072092,
|
| 2325 |
+
"eval_loss": 0.4641817808151245,
|
| 2326 |
+
"eval_runtime": 138.2726,
|
| 2327 |
+
"eval_samples_per_second": 128.456,
|
| 2328 |
+
"eval_steps_per_second": 4.021,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 22.898368883312422,
|
| 2333 |
+
"grad_norm": 1.6573147773742676,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.4327,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 22.898368883312422,
|
| 2340 |
+
"eval_accuracy": 0.8978617589381044,
|
| 2341 |
+
"eval_loss": 0.4600158929824829,
|
| 2342 |
+
"eval_runtime": 139.7334,
|
| 2343 |
+
"eval_samples_per_second": 127.113,
|
| 2344 |
+
"eval_steps_per_second": 3.979,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 23.055207026348807,
|
| 2349 |
+
"grad_norm": 1.7587124109268188,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.4295,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 23.055207026348807,
|
| 2356 |
+
"eval_accuracy": 0.8983885488794708,
|
| 2357 |
+
"eval_loss": 0.4591095745563507,
|
| 2358 |
+
"eval_runtime": 137.0497,
|
| 2359 |
+
"eval_samples_per_second": 129.603,
|
| 2360 |
+
"eval_steps_per_second": 4.057,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 23.212045169385195,
|
| 2365 |
+
"grad_norm": 1.7489006519317627,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.4238,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 23.212045169385195,
|
| 2372 |
+
"eval_accuracy": 0.8974971009561636,
|
| 2373 |
+
"eval_loss": 0.4666588306427002,
|
| 2374 |
+
"eval_runtime": 136.075,
|
| 2375 |
+
"eval_samples_per_second": 130.531,
|
| 2376 |
+
"eval_steps_per_second": 4.086,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 23.36888331242158,
|
| 2381 |
+
"grad_norm": 1.7246719598770142,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.4275,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 23.36888331242158,
|
| 2388 |
+
"eval_accuracy": 0.8988522817576634,
|
| 2389 |
+
"eval_loss": 0.46227309107780457,
|
| 2390 |
+
"eval_runtime": 136.2028,
|
| 2391 |
+
"eval_samples_per_second": 130.409,
|
| 2392 |
+
"eval_steps_per_second": 4.082,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 23.52572145545797,
|
| 2397 |
+
"grad_norm": 1.7149808406829834,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.4289,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 23.52572145545797,
|
| 2404 |
+
"eval_accuracy": 0.89895165608141,
|
| 2405 |
+
"eval_loss": 0.45433443784713745,
|
| 2406 |
+
"eval_runtime": 136.1435,
|
| 2407 |
+
"eval_samples_per_second": 130.465,
|
| 2408 |
+
"eval_steps_per_second": 4.084,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 23.682559598494354,
|
| 2413 |
+
"grad_norm": 1.6236809492111206,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.4248,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 23.682559598494354,
|
| 2420 |
+
"eval_accuracy": 0.8984724067925737,
|
| 2421 |
+
"eval_loss": 0.4650282561779022,
|
| 2422 |
+
"eval_runtime": 136.1923,
|
| 2423 |
+
"eval_samples_per_second": 130.419,
|
| 2424 |
+
"eval_steps_per_second": 4.082,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 23.83939774153074,
|
| 2429 |
+
"grad_norm": 1.8956775665283203,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.4283,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 23.83939774153074,
|
| 2436 |
+
"eval_accuracy": 0.899145284319805,
|
| 2437 |
+
"eval_loss": 0.4555129110813141,
|
| 2438 |
+
"eval_runtime": 136.3725,
|
| 2439 |
+
"eval_samples_per_second": 130.246,
|
| 2440 |
+
"eval_steps_per_second": 4.077,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 23.996235884567128,
|
| 2445 |
+
"grad_norm": 1.5858089923858643,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.4234,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 23.996235884567128,
|
| 2452 |
+
"eval_accuracy": 0.8990089886061022,
|
| 2453 |
+
"eval_loss": 0.45690953731536865,
|
| 2454 |
+
"eval_runtime": 137.2369,
|
| 2455 |
+
"eval_samples_per_second": 129.426,
|
| 2456 |
+
"eval_steps_per_second": 4.051,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 24.153074027603513,
|
| 2461 |
+
"grad_norm": 1.5642129182815552,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.4252,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 24.153074027603513,
|
| 2468 |
+
"eval_accuracy": 0.8991812370142663,
|
| 2469 |
+
"eval_loss": 0.4567597806453705,
|
| 2470 |
+
"eval_runtime": 141.5056,
|
| 2471 |
+
"eval_samples_per_second": 125.522,
|
| 2472 |
+
"eval_steps_per_second": 3.929,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 24.3099121706399,
|
| 2477 |
+
"grad_norm": 1.5789284706115723,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.422,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 24.3099121706399,
|
| 2484 |
+
"eval_accuracy": 0.8998288035249293,
|
| 2485 |
+
"eval_loss": 0.4559558033943176,
|
| 2486 |
+
"eval_runtime": 136.9706,
|
| 2487 |
+
"eval_samples_per_second": 129.677,
|
| 2488 |
+
"eval_steps_per_second": 4.059,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 24.466750313676286,
|
| 2493 |
+
"grad_norm": 1.8032532930374146,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.4206,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 24.466750313676286,
|
| 2500 |
+
"eval_accuracy": 0.8994397460158801,
|
| 2501 |
+
"eval_loss": 0.4534006416797638,
|
| 2502 |
+
"eval_runtime": 136.0321,
|
| 2503 |
+
"eval_samples_per_second": 130.572,
|
| 2504 |
+
"eval_steps_per_second": 4.087,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 24.62358845671267,
|
| 2509 |
+
"grad_norm": 1.7652056217193604,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.4224,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 24.62358845671267,
|
| 2516 |
+
"eval_accuracy": 0.8999375901102459,
|
| 2517 |
+
"eval_loss": 0.4525335133075714,
|
| 2518 |
+
"eval_runtime": 136.0701,
|
| 2519 |
+
"eval_samples_per_second": 130.536,
|
| 2520 |
+
"eval_steps_per_second": 4.086,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 24.78042659974906,
|
| 2525 |
+
"grad_norm": 1.885223388671875,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.4139,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 24.78042659974906,
|
| 2532 |
+
"eval_accuracy": 0.8996436415874084,
|
| 2533 |
+
"eval_loss": 0.45650288462638855,
|
| 2534 |
+
"eval_runtime": 136.924,
|
| 2535 |
+
"eval_samples_per_second": 129.722,
|
| 2536 |
+
"eval_steps_per_second": 4.061,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 24.937264742785445,
|
| 2541 |
+
"grad_norm": 2.1714119911193848,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.4175,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 24.937264742785445,
|
| 2548 |
+
"eval_accuracy": 0.90042263164663,
|
| 2549 |
+
"eval_loss": 0.4520851671695709,
|
| 2550 |
+
"eval_runtime": 135.8737,
|
| 2551 |
+
"eval_samples_per_second": 130.724,
|
| 2552 |
+
"eval_steps_per_second": 4.092,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 25.094102885821833,
|
| 2557 |
+
"grad_norm": 1.8130502700805664,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.4133,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 25.094102885821833,
|
| 2564 |
+
"eval_accuracy": 0.9000717703141665,
|
| 2565 |
+
"eval_loss": 0.4481905400753021,
|
| 2566 |
+
"eval_runtime": 135.7637,
|
| 2567 |
+
"eval_samples_per_second": 130.83,
|
| 2568 |
+
"eval_steps_per_second": 4.095,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 25.250941028858218,
|
| 2573 |
+
"grad_norm": 1.6865718364715576,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.411,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 25.250941028858218,
|
| 2580 |
+
"eval_accuracy": 0.8995766698024459,
|
| 2581 |
+
"eval_loss": 0.4528583288192749,
|
| 2582 |
+
"eval_runtime": 135.904,
|
| 2583 |
+
"eval_samples_per_second": 130.695,
|
| 2584 |
+
"eval_steps_per_second": 4.091,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 25.407779171894603,
|
| 2589 |
+
"grad_norm": 1.7092443704605103,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.4118,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 25.407779171894603,
|
| 2596 |
+
"eval_accuracy": 0.8998159640339154,
|
| 2597 |
+
"eval_loss": 0.457785964012146,
|
| 2598 |
+
"eval_runtime": 136.2219,
|
| 2599 |
+
"eval_samples_per_second": 130.39,
|
| 2600 |
+
"eval_steps_per_second": 4.082,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 25.56461731493099,
|
| 2605 |
+
"grad_norm": 1.8535442352294922,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.4129,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 25.56461731493099,
|
| 2612 |
+
"eval_accuracy": 0.900813356684504,
|
| 2613 |
+
"eval_loss": 0.44758960604667664,
|
| 2614 |
+
"eval_runtime": 139.2861,
|
| 2615 |
+
"eval_samples_per_second": 127.522,
|
| 2616 |
+
"eval_steps_per_second": 3.992,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 25.721455457967377,
|
| 2621 |
+
"grad_norm": 1.7086068391799927,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.4121,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 25.721455457967377,
|
| 2628 |
+
"eval_accuracy": 0.9004114227249277,
|
| 2629 |
+
"eval_loss": 0.4506523609161377,
|
| 2630 |
+
"eval_runtime": 138.1569,
|
| 2631 |
+
"eval_samples_per_second": 128.564,
|
| 2632 |
+
"eval_steps_per_second": 4.024,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 25.878293601003765,
|
| 2637 |
+
"grad_norm": 1.6297106742858887,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.4154,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 25.878293601003765,
|
| 2644 |
+
"eval_accuracy": 0.9013617562120861,
|
| 2645 |
+
"eval_loss": 0.4431464672088623,
|
| 2646 |
+
"eval_runtime": 139.0397,
|
| 2647 |
+
"eval_samples_per_second": 127.748,
|
| 2648 |
+
"eval_steps_per_second": 3.999,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 26.03513174404015,
|
| 2653 |
+
"grad_norm": 1.5048043727874756,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.4122,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 26.03513174404015,
|
| 2660 |
+
"eval_accuracy": 0.9011033099297894,
|
| 2661 |
+
"eval_loss": 0.44738340377807617,
|
| 2662 |
+
"eval_runtime": 138.1953,
|
| 2663 |
+
"eval_samples_per_second": 128.528,
|
| 2664 |
+
"eval_steps_per_second": 4.023,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 26.19196988707654,
|
| 2669 |
+
"grad_norm": 1.5389457941055298,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.4039,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 26.19196988707654,
|
| 2676 |
+
"eval_accuracy": 0.9013423290707173,
|
| 2677 |
+
"eval_loss": 0.4484352171421051,
|
| 2678 |
+
"eval_runtime": 139.444,
|
| 2679 |
+
"eval_samples_per_second": 127.377,
|
| 2680 |
+
"eval_steps_per_second": 3.987,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 26.348808030112924,
|
| 2685 |
+
"grad_norm": 1.6823935508728027,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.4088,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 26.348808030112924,
|
| 2692 |
+
"eval_accuracy": 0.9010903142793143,
|
| 2693 |
+
"eval_loss": 0.44959020614624023,
|
| 2694 |
+
"eval_runtime": 137.4687,
|
| 2695 |
+
"eval_samples_per_second": 129.208,
|
| 2696 |
+
"eval_steps_per_second": 4.045,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 26.50564617314931,
|
| 2701 |
+
"grad_norm": 2.0916717052459717,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.4087,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 26.50564617314931,
|
| 2708 |
+
"eval_accuracy": 0.9014141879881123,
|
| 2709 |
+
"eval_loss": 0.4489114582538605,
|
| 2710 |
+
"eval_runtime": 141.64,
|
| 2711 |
+
"eval_samples_per_second": 125.402,
|
| 2712 |
+
"eval_steps_per_second": 3.925,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 26.662484316185697,
|
| 2717 |
+
"grad_norm": 1.6017004251480103,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.407,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 26.662484316185697,
|
| 2724 |
+
"eval_accuracy": 0.9018399585664477,
|
| 2725 |
+
"eval_loss": 0.4428856372833252,
|
| 2726 |
+
"eval_runtime": 139.1664,
|
| 2727 |
+
"eval_samples_per_second": 127.631,
|
| 2728 |
+
"eval_steps_per_second": 3.995,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 26.819322459222082,
|
| 2733 |
+
"grad_norm": 1.6801029443740845,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.4084,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 26.819322459222082,
|
| 2740 |
+
"eval_accuracy": 0.9015455333911535,
|
| 2741 |
+
"eval_loss": 0.4407350718975067,
|
| 2742 |
+
"eval_runtime": 137.0061,
|
| 2743 |
+
"eval_samples_per_second": 129.644,
|
| 2744 |
+
"eval_steps_per_second": 4.058,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 26.97616060225847,
|
| 2749 |
+
"grad_norm": 1.573623538017273,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.4089,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 26.97616060225847,
|
| 2756 |
+
"eval_accuracy": 0.9024462112584997,
|
| 2757 |
+
"eval_loss": 0.4383965730667114,
|
| 2758 |
+
"eval_runtime": 136.7324,
|
| 2759 |
+
"eval_samples_per_second": 129.903,
|
| 2760 |
+
"eval_steps_per_second": 4.066,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 27.132998745294856,
|
| 2765 |
+
"grad_norm": 1.8882684707641602,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.4044,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 27.132998745294856,
|
| 2772 |
+
"eval_accuracy": 0.9015809796520456,
|
| 2773 |
+
"eval_loss": 0.44848066568374634,
|
| 2774 |
+
"eval_runtime": 137.3268,
|
| 2775 |
+
"eval_samples_per_second": 129.341,
|
| 2776 |
+
"eval_steps_per_second": 4.049,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 27.28983688833124,
|
| 2781 |
+
"grad_norm": 1.6625254154205322,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.4039,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 27.28983688833124,
|
| 2788 |
+
"eval_accuracy": 0.90228319767336,
|
| 2789 |
+
"eval_loss": 0.4422359764575958,
|
| 2790 |
+
"eval_runtime": 137.6608,
|
| 2791 |
+
"eval_samples_per_second": 129.027,
|
| 2792 |
+
"eval_steps_per_second": 4.039,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 27.44667503136763,
|
| 2797 |
+
"grad_norm": 1.8694974184036255,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.4047,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 27.44667503136763,
|
| 2804 |
+
"eval_accuracy": 0.9021538338728836,
|
| 2805 |
+
"eval_loss": 0.4425004720687866,
|
| 2806 |
+
"eval_runtime": 136.387,
|
| 2807 |
+
"eval_samples_per_second": 130.232,
|
| 2808 |
+
"eval_steps_per_second": 4.077,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 27.603513174404014,
|
| 2813 |
+
"grad_norm": 1.5310094356536865,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.405,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 27.603513174404014,
|
| 2820 |
+
"eval_accuracy": 0.9024052445285714,
|
| 2821 |
+
"eval_loss": 0.44192618131637573,
|
| 2822 |
+
"eval_runtime": 135.5046,
|
| 2823 |
+
"eval_samples_per_second": 131.08,
|
| 2824 |
+
"eval_steps_per_second": 4.103,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 27.760351317440403,
|
| 2829 |
+
"grad_norm": 1.5891153812408447,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.4012,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 27.760351317440403,
|
| 2836 |
+
"eval_accuracy": 0.9027056964914371,
|
| 2837 |
+
"eval_loss": 0.44117411971092224,
|
| 2838 |
+
"eval_runtime": 135.8853,
|
| 2839 |
+
"eval_samples_per_second": 130.713,
|
| 2840 |
+
"eval_steps_per_second": 4.092,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 27.917189460476788,
|
| 2845 |
+
"grad_norm": 1.7761974334716797,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.4018,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 27.917189460476788,
|
| 2852 |
+
"eval_accuracy": 0.9025191618761976,
|
| 2853 |
+
"eval_loss": 0.44115549325942993,
|
| 2854 |
+
"eval_runtime": 136.0292,
|
| 2855 |
+
"eval_samples_per_second": 130.575,
|
| 2856 |
+
"eval_steps_per_second": 4.087,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 28.074027603513173,
|
| 2861 |
+
"grad_norm": 1.9455006122589111,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.4008,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 28.074027603513173,
|
| 2868 |
+
"eval_accuracy": 0.9029322223167752,
|
| 2869 |
+
"eval_loss": 0.43775802850723267,
|
| 2870 |
+
"eval_runtime": 135.7499,
|
| 2871 |
+
"eval_samples_per_second": 130.844,
|
| 2872 |
+
"eval_steps_per_second": 4.096,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 28.23086574654956,
|
| 2877 |
+
"grad_norm": 1.6156623363494873,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.3971,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 28.23086574654956,
|
| 2884 |
+
"eval_accuracy": 0.9031078245748608,
|
| 2885 |
+
"eval_loss": 0.43474775552749634,
|
| 2886 |
+
"eval_runtime": 136.1423,
|
| 2887 |
+
"eval_samples_per_second": 130.466,
|
| 2888 |
+
"eval_steps_per_second": 4.084,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 28.387703889585946,
|
| 2893 |
+
"grad_norm": 2.0340116024017334,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.4014,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 28.387703889585946,
|
| 2900 |
+
"eval_accuracy": 0.902757969971607,
|
| 2901 |
+
"eval_loss": 0.4395648241043091,
|
| 2902 |
+
"eval_runtime": 137.427,
|
| 2903 |
+
"eval_samples_per_second": 129.247,
|
| 2904 |
+
"eval_steps_per_second": 4.046,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 28.544542032622335,
|
| 2909 |
+
"grad_norm": 2.1368916034698486,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.3992,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 28.544542032622335,
|
| 2916 |
+
"eval_accuracy": 0.9034885174408511,
|
| 2917 |
+
"eval_loss": 0.4384869635105133,
|
| 2918 |
+
"eval_runtime": 140.4455,
|
| 2919 |
+
"eval_samples_per_second": 126.469,
|
| 2920 |
+
"eval_steps_per_second": 3.959,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 28.70138017565872,
|
| 2925 |
+
"grad_norm": 1.733713150024414,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.3979,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 28.70138017565872,
|
| 2932 |
+
"eval_accuracy": 0.9037653488567973,
|
| 2933 |
+
"eval_loss": 0.4333880841732025,
|
| 2934 |
+
"eval_runtime": 136.0654,
|
| 2935 |
+
"eval_samples_per_second": 130.54,
|
| 2936 |
+
"eval_steps_per_second": 4.086,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 28.85821831869511,
|
| 2941 |
+
"grad_norm": 1.7646188735961914,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.397,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 28.85821831869511,
|
| 2948 |
+
"eval_accuracy": 0.9035005427601347,
|
| 2949 |
+
"eval_loss": 0.43513259291648865,
|
| 2950 |
+
"eval_runtime": 136.1218,
|
| 2951 |
+
"eval_samples_per_second": 130.486,
|
| 2952 |
+
"eval_steps_per_second": 4.085,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 29.015056461731493,
|
| 2957 |
+
"grad_norm": 1.660196304321289,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.3958,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 29.015056461731493,
|
| 2964 |
+
"eval_accuracy": 0.9036445570191227,
|
| 2965 |
+
"eval_loss": 0.4391912519931793,
|
| 2966 |
+
"eval_runtime": 139.306,
|
| 2967 |
+
"eval_samples_per_second": 127.503,
|
| 2968 |
+
"eval_steps_per_second": 3.991,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 29.17189460476788,
|
| 2973 |
+
"grad_norm": 1.5088444948196411,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.3986,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 29.17189460476788,
|
| 2980 |
+
"eval_accuracy": 0.9034735593826558,
|
| 2981 |
+
"eval_loss": 0.4366515874862671,
|
| 2982 |
+
"eval_runtime": 139.2015,
|
| 2983 |
+
"eval_samples_per_second": 127.599,
|
| 2984 |
+
"eval_steps_per_second": 3.994,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 29.328732747804267,
|
| 2989 |
+
"grad_norm": 1.8801556825637817,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.3922,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 29.328732747804267,
|
| 2996 |
+
"eval_accuracy": 0.9034608505271339,
|
| 2997 |
+
"eval_loss": 0.4356890320777893,
|
| 2998 |
+
"eval_runtime": 137.3538,
|
| 2999 |
+
"eval_samples_per_second": 129.316,
|
| 3000 |
+
"eval_steps_per_second": 4.048,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 29.485570890840652,
|
| 3005 |
+
"grad_norm": 1.6091161966323853,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.3921,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 29.485570890840652,
|
| 3012 |
+
"eval_accuracy": 0.9039331120272923,
|
| 3013 |
+
"eval_loss": 0.43036332726478577,
|
| 3014 |
+
"eval_runtime": 136.3616,
|
| 3015 |
+
"eval_samples_per_second": 130.257,
|
| 3016 |
+
"eval_steps_per_second": 4.077,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
}
|
| 3019 |
+
],
|
| 3020 |
+
"logging_steps": 500,
|
| 3021 |
+
"max_steps": 100000,
|
| 3022 |
+
"num_input_tokens_seen": 0,
|
| 3023 |
+
"num_train_epochs": 32,
|
| 3024 |
+
"save_steps": 500,
|
| 3025 |
+
"stateful_callbacks": {
|
| 3026 |
+
"TrainerControl": {
|
| 3027 |
+
"args": {
|
| 3028 |
+
"should_epoch_stop": false,
|
| 3029 |
+
"should_evaluate": false,
|
| 3030 |
+
"should_log": false,
|
| 3031 |
+
"should_save": true,
|
| 3032 |
+
"should_training_stop": false
|
| 3033 |
+
},
|
| 3034 |
+
"attributes": {}
|
| 3035 |
+
}
|
| 3036 |
+
},
|
| 3037 |
+
"total_flos": 7.925427023013478e+17,
|
| 3038 |
+
"train_batch_size": 16,
|
| 3039 |
+
"trial_name": null,
|
| 3040 |
+
"trial_params": null
|
| 3041 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c061aef2e4c01f213384b51a1e999acdc0bb2d55bfefea5c255550ef63f3f29
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|