Uploading checkpoint-99000 for mbert - mlt-latn
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 +3201 -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: mlt-Latn
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# mlt-Latn
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [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: 0.4921
|
| 22 |
+
- Accuracy: 0.9162
|
| 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:fd5ce4c7f4ce475dbd538d3d2c516e48ea6814521011907557b3f979ed29221e
|
| 3 |
+
size 711921812
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:554f5a49e355fffcac910b7f7c20036778023a111eee673676095a6d93f8b28d
|
| 3 |
+
size 1423965114
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bc31a03758ad3846bd5d8236bd270fa8c9867a7295108ec04a7d4a7f3121c7b
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cfd84d77c6613698767788df6e08b5d768bf445bf48de23936380752473535c
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4415404b182ca8f419ce636856a8efa57ef9cce9de682e39e40579cf7bbd5cfa
|
| 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 +1,3201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4882655441761017,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/mbert/mlt-Latn/checkpoint-99000",
|
| 4 |
+
"epoch": 45.020463847203274,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 99000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.22737608003638018,
|
| 13 |
+
"grad_norm": 3.6456298828125,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.6262,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.22737608003638018,
|
| 20 |
+
"eval_accuracy": 0.7730853426906609,
|
| 21 |
+
"eval_loss": 1.2649915218353271,
|
| 22 |
+
"eval_runtime": 31.6229,
|
| 23 |
+
"eval_samples_per_second": 481.866,
|
| 24 |
+
"eval_steps_per_second": 15.084,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.45475216007276037,
|
| 29 |
+
"grad_norm": 4.187455177307129,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.1501,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.45475216007276037,
|
| 36 |
+
"eval_accuracy": 0.8055056158054477,
|
| 37 |
+
"eval_loss": 1.071039080619812,
|
| 38 |
+
"eval_runtime": 32.4866,
|
| 39 |
+
"eval_samples_per_second": 469.055,
|
| 40 |
+
"eval_steps_per_second": 14.683,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.6821282401091405,
|
| 45 |
+
"grad_norm": 2.9080049991607666,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.0315,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.6821282401091405,
|
| 52 |
+
"eval_accuracy": 0.8201615883225247,
|
| 53 |
+
"eval_loss": 0.9629034399986267,
|
| 54 |
+
"eval_runtime": 31.5078,
|
| 55 |
+
"eval_samples_per_second": 483.626,
|
| 56 |
+
"eval_steps_per_second": 15.139,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.9095043201455207,
|
| 61 |
+
"grad_norm": 3.5534873008728027,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 0.9623,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.9095043201455207,
|
| 68 |
+
"eval_accuracy": 0.827454282848158,
|
| 69 |
+
"eval_loss": 0.9285168647766113,
|
| 70 |
+
"eval_runtime": 32.5186,
|
| 71 |
+
"eval_samples_per_second": 468.593,
|
| 72 |
+
"eval_steps_per_second": 14.669,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.1368804001819008,
|
| 77 |
+
"grad_norm": 3.041624069213867,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 0.9065,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1368804001819008,
|
| 84 |
+
"eval_accuracy": 0.8343589346646799,
|
| 85 |
+
"eval_loss": 0.9026671051979065,
|
| 86 |
+
"eval_runtime": 31.4985,
|
| 87 |
+
"eval_samples_per_second": 483.769,
|
| 88 |
+
"eval_steps_per_second": 15.144,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.364256480218281,
|
| 93 |
+
"grad_norm": 3.3451454639434814,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 0.8696,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.364256480218281,
|
| 100 |
+
"eval_accuracy": 0.8405979632481596,
|
| 101 |
+
"eval_loss": 0.8512313365936279,
|
| 102 |
+
"eval_runtime": 32.5429,
|
| 103 |
+
"eval_samples_per_second": 468.243,
|
| 104 |
+
"eval_steps_per_second": 14.658,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.591632560254661,
|
| 109 |
+
"grad_norm": 3.4678544998168945,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 0.8436,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.591632560254661,
|
| 116 |
+
"eval_accuracy": 0.8461289597261696,
|
| 117 |
+
"eval_loss": 0.8300595879554749,
|
| 118 |
+
"eval_runtime": 32.5964,
|
| 119 |
+
"eval_samples_per_second": 467.475,
|
| 120 |
+
"eval_steps_per_second": 14.634,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.8190086402910413,
|
| 125 |
+
"grad_norm": 21.780845642089844,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 0.8104,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.8190086402910413,
|
| 132 |
+
"eval_accuracy": 0.8489302505624955,
|
| 133 |
+
"eval_loss": 0.798919141292572,
|
| 134 |
+
"eval_runtime": 32.5392,
|
| 135 |
+
"eval_samples_per_second": 468.297,
|
| 136 |
+
"eval_steps_per_second": 14.659,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0463847203274215,
|
| 141 |
+
"grad_norm": 3.599536895751953,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 0.7908,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.0463847203274215,
|
| 148 |
+
"eval_accuracy": 0.8513218275815332,
|
| 149 |
+
"eval_loss": 0.8141258955001831,
|
| 150 |
+
"eval_runtime": 34.7129,
|
| 151 |
+
"eval_samples_per_second": 438.972,
|
| 152 |
+
"eval_steps_per_second": 13.741,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.2737608003638017,
|
| 157 |
+
"grad_norm": 3.653501272201538,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 0.764,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.2737608003638017,
|
| 164 |
+
"eval_accuracy": 0.8543282540646533,
|
| 165 |
+
"eval_loss": 0.797082781791687,
|
| 166 |
+
"eval_runtime": 34.8873,
|
| 167 |
+
"eval_samples_per_second": 436.778,
|
| 168 |
+
"eval_steps_per_second": 13.673,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.501136880400182,
|
| 173 |
+
"grad_norm": 3.659949541091919,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 0.7593,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.501136880400182,
|
| 180 |
+
"eval_accuracy": 0.856360096585786,
|
| 181 |
+
"eval_loss": 0.7725093960762024,
|
| 182 |
+
"eval_runtime": 34.9858,
|
| 183 |
+
"eval_samples_per_second": 435.548,
|
| 184 |
+
"eval_steps_per_second": 13.634,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.728512960436562,
|
| 189 |
+
"grad_norm": 2.9741833209991455,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 0.746,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.728512960436562,
|
| 196 |
+
"eval_accuracy": 0.858914082355649,
|
| 197 |
+
"eval_loss": 0.7522701621055603,
|
| 198 |
+
"eval_runtime": 34.645,
|
| 199 |
+
"eval_samples_per_second": 439.832,
|
| 200 |
+
"eval_steps_per_second": 13.768,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.9558890404729423,
|
| 205 |
+
"grad_norm": 4.469085216522217,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 0.7375,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.9558890404729423,
|
| 212 |
+
"eval_accuracy": 0.859327207247945,
|
| 213 |
+
"eval_loss": 0.7622689008712769,
|
| 214 |
+
"eval_runtime": 34.9525,
|
| 215 |
+
"eval_samples_per_second": 435.963,
|
| 216 |
+
"eval_steps_per_second": 13.647,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 3.1832651205093225,
|
| 221 |
+
"grad_norm": 2.7395009994506836,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.7097,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.1832651205093225,
|
| 228 |
+
"eval_accuracy": 0.8622033171867798,
|
| 229 |
+
"eval_loss": 0.747955858707428,
|
| 230 |
+
"eval_runtime": 34.3537,
|
| 231 |
+
"eval_samples_per_second": 443.562,
|
| 232 |
+
"eval_steps_per_second": 13.885,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.4106412005457027,
|
| 237 |
+
"grad_norm": 2.673288106918335,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.7099,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.4106412005457027,
|
| 244 |
+
"eval_accuracy": 0.865153606791071,
|
| 245 |
+
"eval_loss": 0.7201125025749207,
|
| 246 |
+
"eval_runtime": 34.8864,
|
| 247 |
+
"eval_samples_per_second": 436.789,
|
| 248 |
+
"eval_steps_per_second": 13.673,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.6380172805820825,
|
| 253 |
+
"grad_norm": 3.1898393630981445,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.696,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.6380172805820825,
|
| 260 |
+
"eval_accuracy": 0.8656054006196005,
|
| 261 |
+
"eval_loss": 0.7262857556343079,
|
| 262 |
+
"eval_runtime": 33.8112,
|
| 263 |
+
"eval_samples_per_second": 450.679,
|
| 264 |
+
"eval_steps_per_second": 14.108,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 3.865393360618463,
|
| 269 |
+
"grad_norm": 4.6935529708862305,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.6999,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.865393360618463,
|
| 276 |
+
"eval_accuracy": 0.8663027154541763,
|
| 277 |
+
"eval_loss": 0.7238621115684509,
|
| 278 |
+
"eval_runtime": 34.0689,
|
| 279 |
+
"eval_samples_per_second": 447.271,
|
| 280 |
+
"eval_steps_per_second": 14.001,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 4.092769440654843,
|
| 285 |
+
"grad_norm": 3.132990598678589,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.6801,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.092769440654843,
|
| 292 |
+
"eval_accuracy": 0.8682838161725674,
|
| 293 |
+
"eval_loss": 0.7101739048957825,
|
| 294 |
+
"eval_runtime": 33.8325,
|
| 295 |
+
"eval_samples_per_second": 450.395,
|
| 296 |
+
"eval_steps_per_second": 14.099,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.320145520691224,
|
| 301 |
+
"grad_norm": 3.1149518489837646,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.6675,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.320145520691224,
|
| 308 |
+
"eval_accuracy": 0.8696700816141866,
|
| 309 |
+
"eval_loss": 0.7028376460075378,
|
| 310 |
+
"eval_runtime": 34.8653,
|
| 311 |
+
"eval_samples_per_second": 437.054,
|
| 312 |
+
"eval_steps_per_second": 13.681,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 4.547521600727603,
|
| 317 |
+
"grad_norm": 3.1896743774414062,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.6587,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.547521600727603,
|
| 324 |
+
"eval_accuracy": 0.8709308192931519,
|
| 325 |
+
"eval_loss": 0.6970449686050415,
|
| 326 |
+
"eval_runtime": 34.1547,
|
| 327 |
+
"eval_samples_per_second": 446.147,
|
| 328 |
+
"eval_steps_per_second": 13.966,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 4.774897680763983,
|
| 333 |
+
"grad_norm": 2.4510085582733154,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.6565,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.774897680763983,
|
| 340 |
+
"eval_accuracy": 0.8713051922493725,
|
| 341 |
+
"eval_loss": 0.6987766623497009,
|
| 342 |
+
"eval_runtime": 34.6354,
|
| 343 |
+
"eval_samples_per_second": 439.955,
|
| 344 |
+
"eval_steps_per_second": 13.772,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.002273760800364,
|
| 349 |
+
"grad_norm": 2.4753236770629883,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.6564,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 5.002273760800364,
|
| 356 |
+
"eval_accuracy": 0.8719450544496751,
|
| 357 |
+
"eval_loss": 0.6877201795578003,
|
| 358 |
+
"eval_runtime": 34.9052,
|
| 359 |
+
"eval_samples_per_second": 436.553,
|
| 360 |
+
"eval_steps_per_second": 13.666,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 5.229649840836744,
|
| 365 |
+
"grad_norm": 2.689784526824951,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.6294,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.229649840836744,
|
| 372 |
+
"eval_accuracy": 0.8725466807654614,
|
| 373 |
+
"eval_loss": 0.7016832828521729,
|
| 374 |
+
"eval_runtime": 34.9443,
|
| 375 |
+
"eval_samples_per_second": 436.065,
|
| 376 |
+
"eval_steps_per_second": 13.65,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 5.457025920873124,
|
| 381 |
+
"grad_norm": 2.5944247245788574,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.6438,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.457025920873124,
|
| 388 |
+
"eval_accuracy": 0.8738853010432599,
|
| 389 |
+
"eval_loss": 0.6746038198471069,
|
| 390 |
+
"eval_runtime": 34.8136,
|
| 391 |
+
"eval_samples_per_second": 437.702,
|
| 392 |
+
"eval_steps_per_second": 13.702,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 5.684402000909504,
|
| 397 |
+
"grad_norm": 2.799015760421753,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.6266,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 5.684402000909504,
|
| 404 |
+
"eval_accuracy": 0.8748976288586665,
|
| 405 |
+
"eval_loss": 0.6804702877998352,
|
| 406 |
+
"eval_runtime": 34.9715,
|
| 407 |
+
"eval_samples_per_second": 435.726,
|
| 408 |
+
"eval_steps_per_second": 13.64,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 5.911778080945885,
|
| 413 |
+
"grad_norm": 3.0050463676452637,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.6332,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.911778080945885,
|
| 420 |
+
"eval_accuracy": 0.8750289806538035,
|
| 421 |
+
"eval_loss": 0.6853229403495789,
|
| 422 |
+
"eval_runtime": 34.0102,
|
| 423 |
+
"eval_samples_per_second": 448.041,
|
| 424 |
+
"eval_steps_per_second": 14.025,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 6.139154160982264,
|
| 429 |
+
"grad_norm": 3.1468307971954346,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.6257,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 6.139154160982264,
|
| 436 |
+
"eval_accuracy": 0.8770626280480457,
|
| 437 |
+
"eval_loss": 0.6783130764961243,
|
| 438 |
+
"eval_runtime": 34.8957,
|
| 439 |
+
"eval_samples_per_second": 436.672,
|
| 440 |
+
"eval_steps_per_second": 13.669,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.366530241018645,
|
| 445 |
+
"grad_norm": 2.485469102859497,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.6112,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.366530241018645,
|
| 452 |
+
"eval_accuracy": 0.8774595335900176,
|
| 453 |
+
"eval_loss": 0.6619518399238586,
|
| 454 |
+
"eval_runtime": 32.7824,
|
| 455 |
+
"eval_samples_per_second": 464.822,
|
| 456 |
+
"eval_steps_per_second": 14.55,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 6.593906321055025,
|
| 461 |
+
"grad_norm": 2.2372210025787354,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.6064,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.593906321055025,
|
| 468 |
+
"eval_accuracy": 0.8783262916018086,
|
| 469 |
+
"eval_loss": 0.660144567489624,
|
| 470 |
+
"eval_runtime": 32.8384,
|
| 471 |
+
"eval_samples_per_second": 464.03,
|
| 472 |
+
"eval_steps_per_second": 14.526,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 6.8212824010914055,
|
| 477 |
+
"grad_norm": 2.4045872688293457,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.6049,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 6.8212824010914055,
|
| 484 |
+
"eval_accuracy": 0.8777994369334928,
|
| 485 |
+
"eval_loss": 0.6703624129295349,
|
| 486 |
+
"eval_runtime": 31.88,
|
| 487 |
+
"eval_samples_per_second": 477.981,
|
| 488 |
+
"eval_steps_per_second": 14.962,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 7.048658481127785,
|
| 493 |
+
"grad_norm": 4.930024147033691,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.6001,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 7.048658481127785,
|
| 500 |
+
"eval_accuracy": 0.8787791891870121,
|
| 501 |
+
"eval_loss": 0.6632522344589233,
|
| 502 |
+
"eval_runtime": 32.8472,
|
| 503 |
+
"eval_samples_per_second": 463.906,
|
| 504 |
+
"eval_steps_per_second": 14.522,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 7.276034561164166,
|
| 509 |
+
"grad_norm": 2.33658504486084,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.5889,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 7.276034561164166,
|
| 516 |
+
"eval_accuracy": 0.8800836374204799,
|
| 517 |
+
"eval_loss": 0.6591240763664246,
|
| 518 |
+
"eval_runtime": 32.827,
|
| 519 |
+
"eval_samples_per_second": 464.191,
|
| 520 |
+
"eval_steps_per_second": 14.531,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 7.503410641200546,
|
| 525 |
+
"grad_norm": 3.9054555892944336,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.5913,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 7.503410641200546,
|
| 532 |
+
"eval_accuracy": 0.8811457687938868,
|
| 533 |
+
"eval_loss": 0.6544212102890015,
|
| 534 |
+
"eval_runtime": 32.8466,
|
| 535 |
+
"eval_samples_per_second": 463.914,
|
| 536 |
+
"eval_steps_per_second": 14.522,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 7.730786721236926,
|
| 541 |
+
"grad_norm": 3.578624963760376,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.5884,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 7.730786721236926,
|
| 548 |
+
"eval_accuracy": 0.8811873686111147,
|
| 549 |
+
"eval_loss": 0.662954568862915,
|
| 550 |
+
"eval_runtime": 32.8583,
|
| 551 |
+
"eval_samples_per_second": 463.749,
|
| 552 |
+
"eval_steps_per_second": 14.517,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 7.958162801273306,
|
| 557 |
+
"grad_norm": 2.392240047454834,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.5821,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 7.958162801273306,
|
| 564 |
+
"eval_accuracy": 0.8821286707334374,
|
| 565 |
+
"eval_loss": 0.6414864659309387,
|
| 566 |
+
"eval_runtime": 31.9212,
|
| 567 |
+
"eval_samples_per_second": 477.363,
|
| 568 |
+
"eval_steps_per_second": 14.943,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.185538881309686,
|
| 573 |
+
"grad_norm": 2.8014936447143555,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.5725,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 8.185538881309686,
|
| 580 |
+
"eval_accuracy": 0.8820199658271615,
|
| 581 |
+
"eval_loss": 0.641411304473877,
|
| 582 |
+
"eval_runtime": 32.9027,
|
| 583 |
+
"eval_samples_per_second": 463.123,
|
| 584 |
+
"eval_steps_per_second": 14.497,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.412914961346067,
|
| 589 |
+
"grad_norm": 2.3196864128112793,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.5687,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.412914961346067,
|
| 596 |
+
"eval_accuracy": 0.882977694843806,
|
| 597 |
+
"eval_loss": 0.6500121355056763,
|
| 598 |
+
"eval_runtime": 35.1075,
|
| 599 |
+
"eval_samples_per_second": 434.039,
|
| 600 |
+
"eval_steps_per_second": 13.587,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 8.640291041382447,
|
| 605 |
+
"grad_norm": 2.0959160327911377,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.5716,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 8.640291041382447,
|
| 612 |
+
"eval_accuracy": 0.8830600846698087,
|
| 613 |
+
"eval_loss": 0.6408029198646545,
|
| 614 |
+
"eval_runtime": 34.9312,
|
| 615 |
+
"eval_samples_per_second": 436.229,
|
| 616 |
+
"eval_steps_per_second": 13.655,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 8.867667121418826,
|
| 621 |
+
"grad_norm": 2.594998598098755,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.5718,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 8.867667121418826,
|
| 628 |
+
"eval_accuracy": 0.8842278415688298,
|
| 629 |
+
"eval_loss": 0.6326043605804443,
|
| 630 |
+
"eval_runtime": 34.6336,
|
| 631 |
+
"eval_samples_per_second": 439.978,
|
| 632 |
+
"eval_steps_per_second": 13.773,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 9.095043201455207,
|
| 637 |
+
"grad_norm": 2.946044445037842,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.5599,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 9.095043201455207,
|
| 644 |
+
"eval_accuracy": 0.8860304024345182,
|
| 645 |
+
"eval_loss": 0.6145840287208557,
|
| 646 |
+
"eval_runtime": 35.1787,
|
| 647 |
+
"eval_samples_per_second": 433.159,
|
| 648 |
+
"eval_steps_per_second": 13.559,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 9.322419281491587,
|
| 653 |
+
"grad_norm": 2.285879373550415,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.5566,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 9.322419281491587,
|
| 660 |
+
"eval_accuracy": 0.8854829698666589,
|
| 661 |
+
"eval_loss": 0.6315300464630127,
|
| 662 |
+
"eval_runtime": 35.0994,
|
| 663 |
+
"eval_samples_per_second": 434.138,
|
| 664 |
+
"eval_steps_per_second": 13.59,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 9.549795361527968,
|
| 669 |
+
"grad_norm": 3.60970401763916,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.5507,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 9.549795361527968,
|
| 676 |
+
"eval_accuracy": 0.8851466600459794,
|
| 677 |
+
"eval_loss": 0.6236722469329834,
|
| 678 |
+
"eval_runtime": 35.2227,
|
| 679 |
+
"eval_samples_per_second": 432.619,
|
| 680 |
+
"eval_steps_per_second": 13.542,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 9.777171441564347,
|
| 685 |
+
"grad_norm": 2.406843423843384,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.5529,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 9.777171441564347,
|
| 692 |
+
"eval_accuracy": 0.8858670546746865,
|
| 693 |
+
"eval_loss": 0.6286270618438721,
|
| 694 |
+
"eval_runtime": 34.99,
|
| 695 |
+
"eval_samples_per_second": 435.496,
|
| 696 |
+
"eval_steps_per_second": 13.632,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 10.004547521600728,
|
| 701 |
+
"grad_norm": 2.198061466217041,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.5535,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 10.004547521600728,
|
| 708 |
+
"eval_accuracy": 0.8862023923992121,
|
| 709 |
+
"eval_loss": 0.6447569131851196,
|
| 710 |
+
"eval_runtime": 34.6457,
|
| 711 |
+
"eval_samples_per_second": 439.824,
|
| 712 |
+
"eval_steps_per_second": 13.768,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 10.231923601637108,
|
| 717 |
+
"grad_norm": 2.630204439163208,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.5462,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 10.231923601637108,
|
| 724 |
+
"eval_accuracy": 0.8875203016662285,
|
| 725 |
+
"eval_loss": 0.6208702921867371,
|
| 726 |
+
"eval_runtime": 34.9263,
|
| 727 |
+
"eval_samples_per_second": 436.29,
|
| 728 |
+
"eval_steps_per_second": 13.657,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 10.459299681673487,
|
| 733 |
+
"grad_norm": 2.8386332988739014,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.5338,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 10.459299681673487,
|
| 740 |
+
"eval_accuracy": 0.8869479591228265,
|
| 741 |
+
"eval_loss": 0.6332632899284363,
|
| 742 |
+
"eval_runtime": 35.2831,
|
| 743 |
+
"eval_samples_per_second": 431.879,
|
| 744 |
+
"eval_steps_per_second": 13.519,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 10.686675761709868,
|
| 749 |
+
"grad_norm": 1.9988718032836914,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.5388,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 10.686675761709868,
|
| 756 |
+
"eval_accuracy": 0.8874601901854517,
|
| 757 |
+
"eval_loss": 0.6201009750366211,
|
| 758 |
+
"eval_runtime": 33.9197,
|
| 759 |
+
"eval_samples_per_second": 449.238,
|
| 760 |
+
"eval_steps_per_second": 14.063,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 10.914051841746248,
|
| 765 |
+
"grad_norm": 1.991021752357483,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.538,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 10.914051841746248,
|
| 772 |
+
"eval_accuracy": 0.8874614337385308,
|
| 773 |
+
"eval_loss": 0.6168850064277649,
|
| 774 |
+
"eval_runtime": 34.762,
|
| 775 |
+
"eval_samples_per_second": 438.352,
|
| 776 |
+
"eval_steps_per_second": 13.722,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 11.141427921782629,
|
| 781 |
+
"grad_norm": 4.4386491775512695,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.523,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 11.141427921782629,
|
| 788 |
+
"eval_accuracy": 0.8893534276348779,
|
| 789 |
+
"eval_loss": 0.6048386693000793,
|
| 790 |
+
"eval_runtime": 33.7984,
|
| 791 |
+
"eval_samples_per_second": 450.85,
|
| 792 |
+
"eval_steps_per_second": 14.113,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 11.368804001819008,
|
| 797 |
+
"grad_norm": 2.1188158988952637,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.5253,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 11.368804001819008,
|
| 804 |
+
"eval_accuracy": 0.8884587041459525,
|
| 805 |
+
"eval_loss": 0.6149581074714661,
|
| 806 |
+
"eval_runtime": 34.1221,
|
| 807 |
+
"eval_samples_per_second": 446.573,
|
| 808 |
+
"eval_steps_per_second": 13.979,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 11.596180081855389,
|
| 813 |
+
"grad_norm": 4.15702486038208,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.5236,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 11.596180081855389,
|
| 820 |
+
"eval_accuracy": 0.8888612674680825,
|
| 821 |
+
"eval_loss": 0.6105242371559143,
|
| 822 |
+
"eval_runtime": 34.117,
|
| 823 |
+
"eval_samples_per_second": 446.64,
|
| 824 |
+
"eval_steps_per_second": 13.981,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 11.82355616189177,
|
| 829 |
+
"grad_norm": 2.6360466480255127,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.5291,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 11.82355616189177,
|
| 836 |
+
"eval_accuracy": 0.8889885381772583,
|
| 837 |
+
"eval_loss": 0.599651038646698,
|
| 838 |
+
"eval_runtime": 34.8897,
|
| 839 |
+
"eval_samples_per_second": 436.748,
|
| 840 |
+
"eval_steps_per_second": 13.672,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 12.05093224192815,
|
| 845 |
+
"grad_norm": 2.299792766571045,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.5247,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 12.05093224192815,
|
| 852 |
+
"eval_accuracy": 0.8902245874882583,
|
| 853 |
+
"eval_loss": 0.6177428960800171,
|
| 854 |
+
"eval_runtime": 34.0027,
|
| 855 |
+
"eval_samples_per_second": 448.14,
|
| 856 |
+
"eval_steps_per_second": 14.028,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 12.278308321964529,
|
| 861 |
+
"grad_norm": 2.4829318523406982,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.5198,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 12.278308321964529,
|
| 868 |
+
"eval_accuracy": 0.8909571680788114,
|
| 869 |
+
"eval_loss": 0.598882794380188,
|
| 870 |
+
"eval_runtime": 34.9771,
|
| 871 |
+
"eval_samples_per_second": 435.657,
|
| 872 |
+
"eval_steps_per_second": 13.638,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 12.50568440200091,
|
| 877 |
+
"grad_norm": 2.246245861053467,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.5093,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 12.50568440200091,
|
| 884 |
+
"eval_accuracy": 0.8902146740400594,
|
| 885 |
+
"eval_loss": 0.61558598279953,
|
| 886 |
+
"eval_runtime": 35.4475,
|
| 887 |
+
"eval_samples_per_second": 429.876,
|
| 888 |
+
"eval_steps_per_second": 13.457,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 12.73306048203729,
|
| 893 |
+
"grad_norm": 1.9135738611221313,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.5154,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 12.73306048203729,
|
| 900 |
+
"eval_accuracy": 0.8906471410400878,
|
| 901 |
+
"eval_loss": 0.6031416654586792,
|
| 902 |
+
"eval_runtime": 34.9333,
|
| 903 |
+
"eval_samples_per_second": 436.202,
|
| 904 |
+
"eval_steps_per_second": 13.655,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 12.96043656207367,
|
| 909 |
+
"grad_norm": 2.1967339515686035,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.5152,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 12.96043656207367,
|
| 916 |
+
"eval_accuracy": 0.8918997358924995,
|
| 917 |
+
"eval_loss": 0.5982460379600525,
|
| 918 |
+
"eval_runtime": 31.8164,
|
| 919 |
+
"eval_samples_per_second": 478.936,
|
| 920 |
+
"eval_steps_per_second": 14.992,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 13.18781264211005,
|
| 925 |
+
"grad_norm": 2.4718329906463623,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.5045,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 13.18781264211005,
|
| 932 |
+
"eval_accuracy": 0.8919957486399525,
|
| 933 |
+
"eval_loss": 0.6041545271873474,
|
| 934 |
+
"eval_runtime": 31.911,
|
| 935 |
+
"eval_samples_per_second": 477.516,
|
| 936 |
+
"eval_steps_per_second": 14.948,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 13.41518872214643,
|
| 941 |
+
"grad_norm": 3.0977680683135986,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.4962,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 13.41518872214643,
|
| 948 |
+
"eval_accuracy": 0.8928370594159114,
|
| 949 |
+
"eval_loss": 0.6065428256988525,
|
| 950 |
+
"eval_runtime": 32.8507,
|
| 951 |
+
"eval_samples_per_second": 463.856,
|
| 952 |
+
"eval_steps_per_second": 14.52,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 13.642564802182811,
|
| 957 |
+
"grad_norm": 2.383577823638916,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.5073,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 13.642564802182811,
|
| 964 |
+
"eval_accuracy": 0.8931591356868318,
|
| 965 |
+
"eval_loss": 0.5860170722007751,
|
| 966 |
+
"eval_runtime": 32.8261,
|
| 967 |
+
"eval_samples_per_second": 464.204,
|
| 968 |
+
"eval_steps_per_second": 14.531,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 13.86994088221919,
|
| 973 |
+
"grad_norm": 2.574179172515869,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.5017,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 13.86994088221919,
|
| 980 |
+
"eval_accuracy": 0.8917129707220854,
|
| 981 |
+
"eval_loss": 0.6073939204216003,
|
| 982 |
+
"eval_runtime": 33.5273,
|
| 983 |
+
"eval_samples_per_second": 454.495,
|
| 984 |
+
"eval_steps_per_second": 14.227,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 14.09731696225557,
|
| 989 |
+
"grad_norm": 2.23626971244812,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.4962,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 14.09731696225557,
|
| 996 |
+
"eval_accuracy": 0.893840474578457,
|
| 997 |
+
"eval_loss": 0.5841050148010254,
|
| 998 |
+
"eval_runtime": 32.8556,
|
| 999 |
+
"eval_samples_per_second": 463.787,
|
| 1000 |
+
"eval_steps_per_second": 14.518,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 14.324693042291951,
|
| 1005 |
+
"grad_norm": 2.4161667823791504,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.486,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 14.324693042291951,
|
| 1012 |
+
"eval_accuracy": 0.8939059785072527,
|
| 1013 |
+
"eval_loss": 0.5856323838233948,
|
| 1014 |
+
"eval_runtime": 32.8398,
|
| 1015 |
+
"eval_samples_per_second": 464.01,
|
| 1016 |
+
"eval_steps_per_second": 14.525,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 14.552069122328332,
|
| 1021 |
+
"grad_norm": 2.15714168548584,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.4869,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 14.552069122328332,
|
| 1028 |
+
"eval_accuracy": 0.8940220938383814,
|
| 1029 |
+
"eval_loss": 0.6009542942047119,
|
| 1030 |
+
"eval_runtime": 32.8417,
|
| 1031 |
+
"eval_samples_per_second": 463.984,
|
| 1032 |
+
"eval_steps_per_second": 14.524,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 14.77944520236471,
|
| 1037 |
+
"grad_norm": 1.9148741960525513,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.493,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 14.77944520236471,
|
| 1044 |
+
"eval_accuracy": 0.8946031326671489,
|
| 1045 |
+
"eval_loss": 0.5846772789955139,
|
| 1046 |
+
"eval_runtime": 32.8804,
|
| 1047 |
+
"eval_samples_per_second": 463.437,
|
| 1048 |
+
"eval_steps_per_second": 14.507,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 15.006821282401091,
|
| 1053 |
+
"grad_norm": 2.6363611221313477,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.4936,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 15.006821282401091,
|
| 1060 |
+
"eval_accuracy": 0.8942726579365797,
|
| 1061 |
+
"eval_loss": 0.5807018280029297,
|
| 1062 |
+
"eval_runtime": 32.8612,
|
| 1063 |
+
"eval_samples_per_second": 463.708,
|
| 1064 |
+
"eval_steps_per_second": 14.516,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 15.234197362437472,
|
| 1069 |
+
"grad_norm": 17.104581832885742,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.4822,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 15.234197362437472,
|
| 1076 |
+
"eval_accuracy": 0.8951333112695042,
|
| 1077 |
+
"eval_loss": 0.6010898947715759,
|
| 1078 |
+
"eval_runtime": 32.7805,
|
| 1079 |
+
"eval_samples_per_second": 464.85,
|
| 1080 |
+
"eval_steps_per_second": 14.551,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 15.461573442473851,
|
| 1085 |
+
"grad_norm": 8.868905067443848,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.4753,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 15.461573442473851,
|
| 1092 |
+
"eval_accuracy": 0.8953953162756035,
|
| 1093 |
+
"eval_loss": 0.5800424218177795,
|
| 1094 |
+
"eval_runtime": 32.8475,
|
| 1095 |
+
"eval_samples_per_second": 463.901,
|
| 1096 |
+
"eval_steps_per_second": 14.522,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 15.688949522510232,
|
| 1101 |
+
"grad_norm": 2.4379146099090576,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.4803,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 15.688949522510232,
|
| 1108 |
+
"eval_accuracy": 0.8953750241639281,
|
| 1109 |
+
"eval_loss": 0.5949987173080444,
|
| 1110 |
+
"eval_runtime": 32.8126,
|
| 1111 |
+
"eval_samples_per_second": 464.395,
|
| 1112 |
+
"eval_steps_per_second": 14.537,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 15.916325602546612,
|
| 1117 |
+
"grad_norm": 2.3858790397644043,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.4814,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 15.916325602546612,
|
| 1124 |
+
"eval_accuracy": 0.8949742932247506,
|
| 1125 |
+
"eval_loss": 0.5942074656486511,
|
| 1126 |
+
"eval_runtime": 32.8031,
|
| 1127 |
+
"eval_samples_per_second": 464.53,
|
| 1128 |
+
"eval_steps_per_second": 14.541,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 16.143701682582993,
|
| 1133 |
+
"grad_norm": 2.6405398845672607,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.4768,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 16.143701682582993,
|
| 1140 |
+
"eval_accuracy": 0.8950228228780206,
|
| 1141 |
+
"eval_loss": 0.5859512686729431,
|
| 1142 |
+
"eval_runtime": 32.8072,
|
| 1143 |
+
"eval_samples_per_second": 464.472,
|
| 1144 |
+
"eval_steps_per_second": 14.54,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 16.37107776261937,
|
| 1149 |
+
"grad_norm": 2.724952459335327,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.4761,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 16.37107776261937,
|
| 1156 |
+
"eval_accuracy": 0.8954512637706197,
|
| 1157 |
+
"eval_loss": 0.5882732272148132,
|
| 1158 |
+
"eval_runtime": 32.8081,
|
| 1159 |
+
"eval_samples_per_second": 464.458,
|
| 1160 |
+
"eval_steps_per_second": 14.539,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 16.598453842655754,
|
| 1165 |
+
"grad_norm": 2.3671963214874268,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.4707,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 16.598453842655754,
|
| 1172 |
+
"eval_accuracy": 0.8970677263125354,
|
| 1173 |
+
"eval_loss": 0.5809994339942932,
|
| 1174 |
+
"eval_runtime": 32.6484,
|
| 1175 |
+
"eval_samples_per_second": 466.73,
|
| 1176 |
+
"eval_steps_per_second": 14.61,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 16.825829922692133,
|
| 1181 |
+
"grad_norm": 2.2515628337860107,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.4664,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 16.825829922692133,
|
| 1188 |
+
"eval_accuracy": 0.8963799627266053,
|
| 1189 |
+
"eval_loss": 0.5786083936691284,
|
| 1190 |
+
"eval_runtime": 32.6502,
|
| 1191 |
+
"eval_samples_per_second": 466.704,
|
| 1192 |
+
"eval_steps_per_second": 14.609,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 17.053206002728512,
|
| 1197 |
+
"grad_norm": 2.1204442977905273,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.4665,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 17.053206002728512,
|
| 1204 |
+
"eval_accuracy": 0.8963102838280422,
|
| 1205 |
+
"eval_loss": 0.5854199528694153,
|
| 1206 |
+
"eval_runtime": 32.6412,
|
| 1207 |
+
"eval_samples_per_second": 466.833,
|
| 1208 |
+
"eval_steps_per_second": 14.613,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 17.280582082764894,
|
| 1213 |
+
"grad_norm": 2.1250767707824707,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.4595,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 17.280582082764894,
|
| 1220 |
+
"eval_accuracy": 0.897797845220715,
|
| 1221 |
+
"eval_loss": 0.57394939661026,
|
| 1222 |
+
"eval_runtime": 32.6446,
|
| 1223 |
+
"eval_samples_per_second": 466.785,
|
| 1224 |
+
"eval_steps_per_second": 14.612,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 17.507958162801273,
|
| 1229 |
+
"grad_norm": 2.5175702571868896,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.46,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 17.507958162801273,
|
| 1236 |
+
"eval_accuracy": 0.8982157006082142,
|
| 1237 |
+
"eval_loss": 0.5877761840820312,
|
| 1238 |
+
"eval_runtime": 32.647,
|
| 1239 |
+
"eval_samples_per_second": 466.751,
|
| 1240 |
+
"eval_steps_per_second": 14.611,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 17.735334242837652,
|
| 1245 |
+
"grad_norm": 2.2972216606140137,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.4573,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 17.735334242837652,
|
| 1252 |
+
"eval_accuracy": 0.8976428687638632,
|
| 1253 |
+
"eval_loss": 0.5801327228546143,
|
| 1254 |
+
"eval_runtime": 32.6482,
|
| 1255 |
+
"eval_samples_per_second": 466.734,
|
| 1256 |
+
"eval_steps_per_second": 14.61,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 17.962710322874035,
|
| 1261 |
+
"grad_norm": 2.2303497791290283,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.4624,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 17.962710322874035,
|
| 1268 |
+
"eval_accuracy": 0.8979596110936581,
|
| 1269 |
+
"eval_loss": 0.5815677642822266,
|
| 1270 |
+
"eval_runtime": 32.6093,
|
| 1271 |
+
"eval_samples_per_second": 467.29,
|
| 1272 |
+
"eval_steps_per_second": 14.628,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 18.190086402910413,
|
| 1277 |
+
"grad_norm": 2.007495641708374,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.4559,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 18.190086402910413,
|
| 1284 |
+
"eval_accuracy": 0.898133907102813,
|
| 1285 |
+
"eval_loss": 0.5681473612785339,
|
| 1286 |
+
"eval_runtime": 32.6186,
|
| 1287 |
+
"eval_samples_per_second": 467.157,
|
| 1288 |
+
"eval_steps_per_second": 14.624,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 18.417462482946792,
|
| 1293 |
+
"grad_norm": 2.1063549518585205,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.4518,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 18.417462482946792,
|
| 1300 |
+
"eval_accuracy": 0.8984175993824778,
|
| 1301 |
+
"eval_loss": 0.57720547914505,
|
| 1302 |
+
"eval_runtime": 32.6272,
|
| 1303 |
+
"eval_samples_per_second": 467.033,
|
| 1304 |
+
"eval_steps_per_second": 14.62,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 18.644838562983175,
|
| 1309 |
+
"grad_norm": 2.319910764694214,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.4559,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 18.644838562983175,
|
| 1316 |
+
"eval_accuracy": 0.8981402394804274,
|
| 1317 |
+
"eval_loss": 0.5844323635101318,
|
| 1318 |
+
"eval_runtime": 32.6343,
|
| 1319 |
+
"eval_samples_per_second": 466.932,
|
| 1320 |
+
"eval_steps_per_second": 14.617,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 18.872214643019554,
|
| 1325 |
+
"grad_norm": 2.3522610664367676,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.4493,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 18.872214643019554,
|
| 1332 |
+
"eval_accuracy": 0.899032918193942,
|
| 1333 |
+
"eval_loss": 0.5796687006950378,
|
| 1334 |
+
"eval_runtime": 32.6461,
|
| 1335 |
+
"eval_samples_per_second": 466.763,
|
| 1336 |
+
"eval_steps_per_second": 14.611,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 19.099590723055936,
|
| 1341 |
+
"grad_norm": 2.322256088256836,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.443,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 19.099590723055936,
|
| 1348 |
+
"eval_accuracy": 0.8998881179307944,
|
| 1349 |
+
"eval_loss": 0.569625198841095,
|
| 1350 |
+
"eval_runtime": 32.6352,
|
| 1351 |
+
"eval_samples_per_second": 466.918,
|
| 1352 |
+
"eval_steps_per_second": 14.616,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 19.326966803092315,
|
| 1357 |
+
"grad_norm": 2.148592233657837,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.4494,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 19.326966803092315,
|
| 1364 |
+
"eval_accuracy": 0.8993213326792998,
|
| 1365 |
+
"eval_loss": 0.56317138671875,
|
| 1366 |
+
"eval_runtime": 32.643,
|
| 1367 |
+
"eval_samples_per_second": 466.807,
|
| 1368 |
+
"eval_steps_per_second": 14.613,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 19.554342883128694,
|
| 1373 |
+
"grad_norm": 2.8056764602661133,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.4405,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 19.554342883128694,
|
| 1380 |
+
"eval_accuracy": 0.8992730163225374,
|
| 1381 |
+
"eval_loss": 0.5784369111061096,
|
| 1382 |
+
"eval_runtime": 32.6779,
|
| 1383 |
+
"eval_samples_per_second": 466.31,
|
| 1384 |
+
"eval_steps_per_second": 14.597,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 19.781718963165076,
|
| 1389 |
+
"grad_norm": 4.026996612548828,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.4456,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 19.781718963165076,
|
| 1396 |
+
"eval_accuracy": 0.8996898687782441,
|
| 1397 |
+
"eval_loss": 0.5666890144348145,
|
| 1398 |
+
"eval_runtime": 32.6164,
|
| 1399 |
+
"eval_samples_per_second": 467.189,
|
| 1400 |
+
"eval_steps_per_second": 14.625,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 20.009095043201455,
|
| 1405 |
+
"grad_norm": 2.2922003269195557,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.4413,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 20.009095043201455,
|
| 1412 |
+
"eval_accuracy": 0.9003280188691405,
|
| 1413 |
+
"eval_loss": 0.5622244477272034,
|
| 1414 |
+
"eval_runtime": 32.6479,
|
| 1415 |
+
"eval_samples_per_second": 466.738,
|
| 1416 |
+
"eval_steps_per_second": 14.61,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 20.236471123237834,
|
| 1421 |
+
"grad_norm": 2.7969603538513184,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.4345,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 20.236471123237834,
|
| 1428 |
+
"eval_accuracy": 0.9004365261875368,
|
| 1429 |
+
"eval_loss": 0.5717702507972717,
|
| 1430 |
+
"eval_runtime": 32.6321,
|
| 1431 |
+
"eval_samples_per_second": 466.963,
|
| 1432 |
+
"eval_steps_per_second": 14.618,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 20.463847203274216,
|
| 1437 |
+
"grad_norm": 2.028287887573242,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.4341,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 20.463847203274216,
|
| 1444 |
+
"eval_accuracy": 0.9004377143446534,
|
| 1445 |
+
"eval_loss": 0.5708958506584167,
|
| 1446 |
+
"eval_runtime": 32.6486,
|
| 1447 |
+
"eval_samples_per_second": 466.727,
|
| 1448 |
+
"eval_steps_per_second": 14.61,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 20.691223283310595,
|
| 1453 |
+
"grad_norm": 2.1190261840820312,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.4355,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 20.691223283310595,
|
| 1460 |
+
"eval_accuracy": 0.9009621428226079,
|
| 1461 |
+
"eval_loss": 0.548612654209137,
|
| 1462 |
+
"eval_runtime": 32.6093,
|
| 1463 |
+
"eval_samples_per_second": 467.29,
|
| 1464 |
+
"eval_steps_per_second": 14.628,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 20.918599363346974,
|
| 1469 |
+
"grad_norm": 2.1443231105804443,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.4348,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 20.918599363346974,
|
| 1476 |
+
"eval_accuracy": 0.9010261908408639,
|
| 1477 |
+
"eval_loss": 0.556687593460083,
|
| 1478 |
+
"eval_runtime": 32.6956,
|
| 1479 |
+
"eval_samples_per_second": 466.057,
|
| 1480 |
+
"eval_steps_per_second": 14.589,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 21.145975443383357,
|
| 1485 |
+
"grad_norm": 1.7599616050720215,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.429,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 21.145975443383357,
|
| 1492 |
+
"eval_accuracy": 0.9013362827863302,
|
| 1493 |
+
"eval_loss": 0.5656816959381104,
|
| 1494 |
+
"eval_runtime": 32.6749,
|
| 1495 |
+
"eval_samples_per_second": 466.352,
|
| 1496 |
+
"eval_steps_per_second": 14.598,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 21.373351523419736,
|
| 1501 |
+
"grad_norm": 2.0747954845428467,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.4286,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 21.373351523419736,
|
| 1508 |
+
"eval_accuracy": 0.9021677014422286,
|
| 1509 |
+
"eval_loss": 0.5610804557800293,
|
| 1510 |
+
"eval_runtime": 32.6839,
|
| 1511 |
+
"eval_samples_per_second": 466.223,
|
| 1512 |
+
"eval_steps_per_second": 14.594,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 21.600727603456118,
|
| 1517 |
+
"grad_norm": 2.359894275665283,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.4286,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 21.600727603456118,
|
| 1524 |
+
"eval_accuracy": 0.9011451398135819,
|
| 1525 |
+
"eval_loss": 0.5557427406311035,
|
| 1526 |
+
"eval_runtime": 32.7147,
|
| 1527 |
+
"eval_samples_per_second": 465.785,
|
| 1528 |
+
"eval_steps_per_second": 14.581,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 21.828103683492497,
|
| 1533 |
+
"grad_norm": 2.622968912124634,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.4283,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 21.828103683492497,
|
| 1540 |
+
"eval_accuracy": 0.9020284253331723,
|
| 1541 |
+
"eval_loss": 0.5703328251838684,
|
| 1542 |
+
"eval_runtime": 32.707,
|
| 1543 |
+
"eval_samples_per_second": 465.894,
|
| 1544 |
+
"eval_steps_per_second": 14.584,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 22.055479763528876,
|
| 1549 |
+
"grad_norm": 2.860867738723755,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.4266,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 22.055479763528876,
|
| 1556 |
+
"eval_accuracy": 0.902576896202834,
|
| 1557 |
+
"eval_loss": 0.5727266669273376,
|
| 1558 |
+
"eval_runtime": 32.7207,
|
| 1559 |
+
"eval_samples_per_second": 465.699,
|
| 1560 |
+
"eval_steps_per_second": 14.578,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 22.282855843565258,
|
| 1565 |
+
"grad_norm": 1.9193904399871826,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.421,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 22.282855843565258,
|
| 1572 |
+
"eval_accuracy": 0.9023543270239306,
|
| 1573 |
+
"eval_loss": 0.558401346206665,
|
| 1574 |
+
"eval_runtime": 32.7214,
|
| 1575 |
+
"eval_samples_per_second": 465.69,
|
| 1576 |
+
"eval_steps_per_second": 14.578,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 22.510231923601637,
|
| 1581 |
+
"grad_norm": 2.298145055770874,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.4177,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 22.510231923601637,
|
| 1588 |
+
"eval_accuracy": 0.9031498029862292,
|
| 1589 |
+
"eval_loss": 0.5540251135826111,
|
| 1590 |
+
"eval_runtime": 32.6923,
|
| 1591 |
+
"eval_samples_per_second": 466.104,
|
| 1592 |
+
"eval_steps_per_second": 14.591,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 22.737608003638016,
|
| 1597 |
+
"grad_norm": 1.8608070611953735,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.416,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 22.737608003638016,
|
| 1604 |
+
"eval_accuracy": 0.902929785480278,
|
| 1605 |
+
"eval_loss": 0.5727468729019165,
|
| 1606 |
+
"eval_runtime": 32.7273,
|
| 1607 |
+
"eval_samples_per_second": 465.604,
|
| 1608 |
+
"eval_steps_per_second": 14.575,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 22.9649840836744,
|
| 1613 |
+
"grad_norm": 1.9308445453643799,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.416,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 22.9649840836744,
|
| 1620 |
+
"eval_accuracy": 0.9035583359641267,
|
| 1621 |
+
"eval_loss": 0.5548630952835083,
|
| 1622 |
+
"eval_runtime": 32.7576,
|
| 1623 |
+
"eval_samples_per_second": 465.175,
|
| 1624 |
+
"eval_steps_per_second": 14.562,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 23.192360163710777,
|
| 1629 |
+
"grad_norm": 2.954545021057129,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.4151,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 23.192360163710777,
|
| 1636 |
+
"eval_accuracy": 0.9031502317839712,
|
| 1637 |
+
"eval_loss": 0.5622453093528748,
|
| 1638 |
+
"eval_runtime": 32.7416,
|
| 1639 |
+
"eval_samples_per_second": 465.401,
|
| 1640 |
+
"eval_steps_per_second": 14.569,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 23.419736243747156,
|
| 1645 |
+
"grad_norm": 2.3451988697052,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.4108,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 23.419736243747156,
|
| 1652 |
+
"eval_accuracy": 0.9029909941282878,
|
| 1653 |
+
"eval_loss": 0.5615450739860535,
|
| 1654 |
+
"eval_runtime": 32.7353,
|
| 1655 |
+
"eval_samples_per_second": 465.491,
|
| 1656 |
+
"eval_steps_per_second": 14.571,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 23.64711232378354,
|
| 1661 |
+
"grad_norm": 2.289078950881958,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.4085,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 23.64711232378354,
|
| 1668 |
+
"eval_accuracy": 0.9037381257773947,
|
| 1669 |
+
"eval_loss": 0.5549082159996033,
|
| 1670 |
+
"eval_runtime": 32.7339,
|
| 1671 |
+
"eval_samples_per_second": 465.511,
|
| 1672 |
+
"eval_steps_per_second": 14.572,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 23.874488403819917,
|
| 1677 |
+
"grad_norm": 2.206704616546631,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.4121,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 23.874488403819917,
|
| 1684 |
+
"eval_accuracy": 0.9035641917801583,
|
| 1685 |
+
"eval_loss": 0.5449928641319275,
|
| 1686 |
+
"eval_runtime": 32.7514,
|
| 1687 |
+
"eval_samples_per_second": 465.262,
|
| 1688 |
+
"eval_steps_per_second": 14.564,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 24.1018644838563,
|
| 1693 |
+
"grad_norm": 1.917070984840393,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.4068,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 24.1018644838563,
|
| 1700 |
+
"eval_accuracy": 0.9043728588740845,
|
| 1701 |
+
"eval_loss": 0.5453593134880066,
|
| 1702 |
+
"eval_runtime": 32.7564,
|
| 1703 |
+
"eval_samples_per_second": 465.191,
|
| 1704 |
+
"eval_steps_per_second": 14.562,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 24.32924056389268,
|
| 1709 |
+
"grad_norm": 2.2474703788757324,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.403,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 24.32924056389268,
|
| 1716 |
+
"eval_accuracy": 0.9039751943993605,
|
| 1717 |
+
"eval_loss": 0.5543568730354309,
|
| 1718 |
+
"eval_runtime": 32.7427,
|
| 1719 |
+
"eval_samples_per_second": 465.386,
|
| 1720 |
+
"eval_steps_per_second": 14.568,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 24.556616643929058,
|
| 1725 |
+
"grad_norm": 2.8877437114715576,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.4044,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 24.556616643929058,
|
| 1732 |
+
"eval_accuracy": 0.9040404610755741,
|
| 1733 |
+
"eval_loss": 0.5543816685676575,
|
| 1734 |
+
"eval_runtime": 32.8189,
|
| 1735 |
+
"eval_samples_per_second": 464.306,
|
| 1736 |
+
"eval_steps_per_second": 14.534,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 24.78399272396544,
|
| 1741 |
+
"grad_norm": 2.6778457164764404,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.3986,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 24.78399272396544,
|
| 1748 |
+
"eval_accuracy": 0.9051992065338192,
|
| 1749 |
+
"eval_loss": 0.5703139901161194,
|
| 1750 |
+
"eval_runtime": 32.8924,
|
| 1751 |
+
"eval_samples_per_second": 463.268,
|
| 1752 |
+
"eval_steps_per_second": 14.502,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 25.01136880400182,
|
| 1757 |
+
"grad_norm": 1.8749839067459106,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.4037,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 25.01136880400182,
|
| 1764 |
+
"eval_accuracy": 0.9054868850104956,
|
| 1765 |
+
"eval_loss": 0.5421493053436279,
|
| 1766 |
+
"eval_runtime": 32.9644,
|
| 1767 |
+
"eval_samples_per_second": 462.256,
|
| 1768 |
+
"eval_steps_per_second": 14.47,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 25.238744884038198,
|
| 1773 |
+
"grad_norm": 2.0783512592315674,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.3953,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 25.238744884038198,
|
| 1780 |
+
"eval_accuracy": 0.9047332246864593,
|
| 1781 |
+
"eval_loss": 0.5418171286582947,
|
| 1782 |
+
"eval_runtime": 32.7741,
|
| 1783 |
+
"eval_samples_per_second": 464.94,
|
| 1784 |
+
"eval_steps_per_second": 14.554,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 25.46612096407458,
|
| 1789 |
+
"grad_norm": 2.809805154800415,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.3955,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 25.46612096407458,
|
| 1796 |
+
"eval_accuracy": 0.9053374070652981,
|
| 1797 |
+
"eval_loss": 0.5434351563453674,
|
| 1798 |
+
"eval_runtime": 32.7589,
|
| 1799 |
+
"eval_samples_per_second": 465.156,
|
| 1800 |
+
"eval_steps_per_second": 14.561,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 25.69349704411096,
|
| 1805 |
+
"grad_norm": 1.9876326322555542,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.3959,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 25.69349704411096,
|
| 1812 |
+
"eval_accuracy": 0.9050791584361013,
|
| 1813 |
+
"eval_loss": 0.5530513525009155,
|
| 1814 |
+
"eval_runtime": 32.7387,
|
| 1815 |
+
"eval_samples_per_second": 465.443,
|
| 1816 |
+
"eval_steps_per_second": 14.57,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 25.920873124147338,
|
| 1821 |
+
"grad_norm": 2.88883638381958,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.3935,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 25.920873124147338,
|
| 1828 |
+
"eval_accuracy": 0.9060817940460243,
|
| 1829 |
+
"eval_loss": 0.5371158719062805,
|
| 1830 |
+
"eval_runtime": 32.7408,
|
| 1831 |
+
"eval_samples_per_second": 465.414,
|
| 1832 |
+
"eval_steps_per_second": 14.569,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 26.14824920418372,
|
| 1837 |
+
"grad_norm": 2.2030739784240723,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.3858,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 26.14824920418372,
|
| 1844 |
+
"eval_accuracy": 0.9057486307794966,
|
| 1845 |
+
"eval_loss": 0.5505759119987488,
|
| 1846 |
+
"eval_runtime": 32.7846,
|
| 1847 |
+
"eval_samples_per_second": 464.792,
|
| 1848 |
+
"eval_steps_per_second": 14.55,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 26.3756252842201,
|
| 1853 |
+
"grad_norm": 2.6154465675354004,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.3855,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 26.3756252842201,
|
| 1860 |
+
"eval_accuracy": 0.9060594819521951,
|
| 1861 |
+
"eval_loss": 0.5361325144767761,
|
| 1862 |
+
"eval_runtime": 32.7611,
|
| 1863 |
+
"eval_samples_per_second": 465.124,
|
| 1864 |
+
"eval_steps_per_second": 14.56,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 26.60300136425648,
|
| 1869 |
+
"grad_norm": 2.7764670848846436,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.3915,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 26.60300136425648,
|
| 1876 |
+
"eval_accuracy": 0.90627844666889,
|
| 1877 |
+
"eval_loss": 0.5417194962501526,
|
| 1878 |
+
"eval_runtime": 32.7608,
|
| 1879 |
+
"eval_samples_per_second": 465.128,
|
| 1880 |
+
"eval_steps_per_second": 14.56,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 26.83037744429286,
|
| 1885 |
+
"grad_norm": 2.538780689239502,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.386,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 26.83037744429286,
|
| 1892 |
+
"eval_accuracy": 0.9057192748802404,
|
| 1893 |
+
"eval_loss": 0.5382261872291565,
|
| 1894 |
+
"eval_runtime": 32.7359,
|
| 1895 |
+
"eval_samples_per_second": 465.483,
|
| 1896 |
+
"eval_steps_per_second": 14.571,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 27.05775352432924,
|
| 1901 |
+
"grad_norm": 2.0034031867980957,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.3903,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 27.05775352432924,
|
| 1908 |
+
"eval_accuracy": 0.9067839467973896,
|
| 1909 |
+
"eval_loss": 0.5416918992996216,
|
| 1910 |
+
"eval_runtime": 32.7241,
|
| 1911 |
+
"eval_samples_per_second": 465.651,
|
| 1912 |
+
"eval_steps_per_second": 14.576,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 27.285129604365622,
|
| 1917 |
+
"grad_norm": 2.290728807449341,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.3855,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 27.285129604365622,
|
| 1924 |
+
"eval_accuracy": 0.9074836366327094,
|
| 1925 |
+
"eval_loss": 0.5252583622932434,
|
| 1926 |
+
"eval_runtime": 32.7646,
|
| 1927 |
+
"eval_samples_per_second": 465.075,
|
| 1928 |
+
"eval_steps_per_second": 14.558,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 27.512505684402,
|
| 1933 |
+
"grad_norm": 2.7191412448883057,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.3843,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 27.512505684402,
|
| 1940 |
+
"eval_accuracy": 0.9073820932473238,
|
| 1941 |
+
"eval_loss": 0.5472291707992554,
|
| 1942 |
+
"eval_runtime": 32.7214,
|
| 1943 |
+
"eval_samples_per_second": 465.69,
|
| 1944 |
+
"eval_steps_per_second": 14.578,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 27.73988176443838,
|
| 1949 |
+
"grad_norm": 1.929368019104004,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.3826,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 27.73988176443838,
|
| 1956 |
+
"eval_accuracy": 0.9078339005862959,
|
| 1957 |
+
"eval_loss": 0.5291460156440735,
|
| 1958 |
+
"eval_runtime": 32.743,
|
| 1959 |
+
"eval_samples_per_second": 465.381,
|
| 1960 |
+
"eval_steps_per_second": 14.568,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 27.967257844474762,
|
| 1965 |
+
"grad_norm": 1.5827338695526123,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.3815,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 27.967257844474762,
|
| 1972 |
+
"eval_accuracy": 0.9073046459144095,
|
| 1973 |
+
"eval_loss": 0.5335054397583008,
|
| 1974 |
+
"eval_runtime": 32.7633,
|
| 1975 |
+
"eval_samples_per_second": 465.094,
|
| 1976 |
+
"eval_steps_per_second": 14.559,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 28.19463392451114,
|
| 1981 |
+
"grad_norm": 2.0227861404418945,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.3718,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 28.19463392451114,
|
| 1988 |
+
"eval_accuracy": 0.9079641018191209,
|
| 1989 |
+
"eval_loss": 0.5228089690208435,
|
| 1990 |
+
"eval_runtime": 32.7482,
|
| 1991 |
+
"eval_samples_per_second": 465.308,
|
| 1992 |
+
"eval_steps_per_second": 14.566,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 28.42201000454752,
|
| 1997 |
+
"grad_norm": 2.1523451805114746,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.3781,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 28.42201000454752,
|
| 2004 |
+
"eval_accuracy": 0.9087774304782733,
|
| 2005 |
+
"eval_loss": 0.5337960720062256,
|
| 2006 |
+
"eval_runtime": 32.7551,
|
| 2007 |
+
"eval_samples_per_second": 465.209,
|
| 2008 |
+
"eval_steps_per_second": 14.563,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 28.649386084583902,
|
| 2013 |
+
"grad_norm": 2.558656930923462,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.374,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 28.649386084583902,
|
| 2020 |
+
"eval_accuracy": 0.907944286819706,
|
| 2021 |
+
"eval_loss": 0.5363012552261353,
|
| 2022 |
+
"eval_runtime": 32.751,
|
| 2023 |
+
"eval_samples_per_second": 465.268,
|
| 2024 |
+
"eval_steps_per_second": 14.564,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 28.87676216462028,
|
| 2029 |
+
"grad_norm": 2.0979561805725098,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.3844,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 28.87676216462028,
|
| 2036 |
+
"eval_accuracy": 0.9085302920838064,
|
| 2037 |
+
"eval_loss": 0.5352813601493835,
|
| 2038 |
+
"eval_runtime": 32.8469,
|
| 2039 |
+
"eval_samples_per_second": 463.91,
|
| 2040 |
+
"eval_steps_per_second": 14.522,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 29.104138244656664,
|
| 2045 |
+
"grad_norm": 2.1332693099975586,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.3718,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 29.104138244656664,
|
| 2052 |
+
"eval_accuracy": 0.9081558247133298,
|
| 2053 |
+
"eval_loss": 0.521395742893219,
|
| 2054 |
+
"eval_runtime": 32.7396,
|
| 2055 |
+
"eval_samples_per_second": 465.43,
|
| 2056 |
+
"eval_steps_per_second": 14.57,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 29.331514324693043,
|
| 2061 |
+
"grad_norm": 2.3449409008026123,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.3689,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 29.331514324693043,
|
| 2068 |
+
"eval_accuracy": 0.908818716090388,
|
| 2069 |
+
"eval_loss": 0.5300734639167786,
|
| 2070 |
+
"eval_runtime": 32.7208,
|
| 2071 |
+
"eval_samples_per_second": 465.697,
|
| 2072 |
+
"eval_steps_per_second": 14.578,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 29.55889040472942,
|
| 2077 |
+
"grad_norm": 2.739234685897827,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.3659,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 29.55889040472942,
|
| 2084 |
+
"eval_accuracy": 0.9088500085480837,
|
| 2085 |
+
"eval_loss": 0.5266396999359131,
|
| 2086 |
+
"eval_runtime": 32.8245,
|
| 2087 |
+
"eval_samples_per_second": 464.226,
|
| 2088 |
+
"eval_steps_per_second": 14.532,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 29.786266484765804,
|
| 2093 |
+
"grad_norm": 1.9931187629699707,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.3699,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 29.786266484765804,
|
| 2100 |
+
"eval_accuracy": 0.9088424483905984,
|
| 2101 |
+
"eval_loss": 0.5349287986755371,
|
| 2102 |
+
"eval_runtime": 32.743,
|
| 2103 |
+
"eval_samples_per_second": 465.382,
|
| 2104 |
+
"eval_steps_per_second": 14.568,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 30.013642564802183,
|
| 2109 |
+
"grad_norm": 2.5198569297790527,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.3722,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 30.013642564802183,
|
| 2116 |
+
"eval_accuracy": 0.9093058727635157,
|
| 2117 |
+
"eval_loss": 0.5210967063903809,
|
| 2118 |
+
"eval_runtime": 32.7556,
|
| 2119 |
+
"eval_samples_per_second": 465.203,
|
| 2120 |
+
"eval_steps_per_second": 14.562,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 30.24101864483856,
|
| 2125 |
+
"grad_norm": 2.4215102195739746,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.3642,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 30.24101864483856,
|
| 2132 |
+
"eval_accuracy": 0.9092281735896366,
|
| 2133 |
+
"eval_loss": 0.5404704809188843,
|
| 2134 |
+
"eval_runtime": 32.7782,
|
| 2135 |
+
"eval_samples_per_second": 464.882,
|
| 2136 |
+
"eval_steps_per_second": 14.552,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 30.468394724874944,
|
| 2141 |
+
"grad_norm": 2.2673470973968506,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.3622,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 30.468394724874944,
|
| 2148 |
+
"eval_accuracy": 0.9083970080404082,
|
| 2149 |
+
"eval_loss": 0.5448651313781738,
|
| 2150 |
+
"eval_runtime": 32.7637,
|
| 2151 |
+
"eval_samples_per_second": 465.088,
|
| 2152 |
+
"eval_steps_per_second": 14.559,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 30.695770804911323,
|
| 2157 |
+
"grad_norm": 2.9746830463409424,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.3629,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 30.695770804911323,
|
| 2164 |
+
"eval_accuracy": 0.9096486227672552,
|
| 2165 |
+
"eval_loss": 0.5271045565605164,
|
| 2166 |
+
"eval_runtime": 32.7536,
|
| 2167 |
+
"eval_samples_per_second": 465.232,
|
| 2168 |
+
"eval_steps_per_second": 14.563,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 30.923146884947702,
|
| 2173 |
+
"grad_norm": 2.3879144191741943,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.3592,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 30.923146884947702,
|
| 2180 |
+
"eval_accuracy": 0.9106144262759255,
|
| 2181 |
+
"eval_loss": 0.5255327820777893,
|
| 2182 |
+
"eval_runtime": 32.7669,
|
| 2183 |
+
"eval_samples_per_second": 465.043,
|
| 2184 |
+
"eval_steps_per_second": 14.557,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 31.150522964984084,
|
| 2189 |
+
"grad_norm": 2.2848832607269287,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.361,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 31.150522964984084,
|
| 2196 |
+
"eval_accuracy": 0.9105147834118348,
|
| 2197 |
+
"eval_loss": 0.5189831256866455,
|
| 2198 |
+
"eval_runtime": 32.733,
|
| 2199 |
+
"eval_samples_per_second": 465.524,
|
| 2200 |
+
"eval_steps_per_second": 14.572,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 31.377899045020463,
|
| 2205 |
+
"grad_norm": 2.483689785003662,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.356,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 31.377899045020463,
|
| 2212 |
+
"eval_accuracy": 0.9096528047653679,
|
| 2213 |
+
"eval_loss": 0.5125781893730164,
|
| 2214 |
+
"eval_runtime": 32.7803,
|
| 2215 |
+
"eval_samples_per_second": 464.853,
|
| 2216 |
+
"eval_steps_per_second": 14.551,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 31.605275125056846,
|
| 2221 |
+
"grad_norm": 3.0858843326568604,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.3568,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 31.605275125056846,
|
| 2228 |
+
"eval_accuracy": 0.9104613248966942,
|
| 2229 |
+
"eval_loss": 0.5205091238021851,
|
| 2230 |
+
"eval_runtime": 32.764,
|
| 2231 |
+
"eval_samples_per_second": 465.083,
|
| 2232 |
+
"eval_steps_per_second": 14.559,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 31.832651205093224,
|
| 2237 |
+
"grad_norm": 2.596628427505493,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.3599,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 31.832651205093224,
|
| 2244 |
+
"eval_accuracy": 0.9102152749211573,
|
| 2245 |
+
"eval_loss": 0.5163892507553101,
|
| 2246 |
+
"eval_runtime": 32.7518,
|
| 2247 |
+
"eval_samples_per_second": 465.256,
|
| 2248 |
+
"eval_steps_per_second": 14.564,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 32.06002728512961,
|
| 2253 |
+
"grad_norm": 1.727830171585083,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.3517,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 32.06002728512961,
|
| 2260 |
+
"eval_accuracy": 0.9108499567185936,
|
| 2261 |
+
"eval_loss": 0.5241507291793823,
|
| 2262 |
+
"eval_runtime": 32.8266,
|
| 2263 |
+
"eval_samples_per_second": 464.197,
|
| 2264 |
+
"eval_steps_per_second": 14.531,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 32.287403365165986,
|
| 2269 |
+
"grad_norm": 2.585827112197876,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.3525,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 32.287403365165986,
|
| 2276 |
+
"eval_accuracy": 0.9102041670961757,
|
| 2277 |
+
"eval_loss": 0.5363503694534302,
|
| 2278 |
+
"eval_runtime": 32.7637,
|
| 2279 |
+
"eval_samples_per_second": 465.087,
|
| 2280 |
+
"eval_steps_per_second": 14.559,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 32.514779445202365,
|
| 2285 |
+
"grad_norm": 1.7349563837051392,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.3519,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 32.514779445202365,
|
| 2292 |
+
"eval_accuracy": 0.9110400082654367,
|
| 2293 |
+
"eval_loss": 0.5117472410202026,
|
| 2294 |
+
"eval_runtime": 32.7979,
|
| 2295 |
+
"eval_samples_per_second": 464.604,
|
| 2296 |
+
"eval_steps_per_second": 14.544,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 32.74215552523874,
|
| 2301 |
+
"grad_norm": 1.795889973640442,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.3477,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 32.74215552523874,
|
| 2308 |
+
"eval_accuracy": 0.9104842102548314,
|
| 2309 |
+
"eval_loss": 0.5326040387153625,
|
| 2310 |
+
"eval_runtime": 32.8076,
|
| 2311 |
+
"eval_samples_per_second": 464.465,
|
| 2312 |
+
"eval_steps_per_second": 14.539,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 32.96953160527512,
|
| 2317 |
+
"grad_norm": 2.50250506401062,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.3521,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 32.96953160527512,
|
| 2324 |
+
"eval_accuracy": 0.910281578904953,
|
| 2325 |
+
"eval_loss": 0.5304139256477356,
|
| 2326 |
+
"eval_runtime": 32.7772,
|
| 2327 |
+
"eval_samples_per_second": 464.896,
|
| 2328 |
+
"eval_steps_per_second": 14.553,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 33.19690768531151,
|
| 2333 |
+
"grad_norm": 1.8461118936538696,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.3437,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 33.19690768531151,
|
| 2340 |
+
"eval_accuracy": 0.9116313736371878,
|
| 2341 |
+
"eval_loss": 0.5194041132926941,
|
| 2342 |
+
"eval_runtime": 32.8126,
|
| 2343 |
+
"eval_samples_per_second": 464.394,
|
| 2344 |
+
"eval_steps_per_second": 14.537,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 33.42428376534789,
|
| 2349 |
+
"grad_norm": 2.4689884185791016,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.3465,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 33.42428376534789,
|
| 2356 |
+
"eval_accuracy": 0.9114267094251479,
|
| 2357 |
+
"eval_loss": 0.520270049571991,
|
| 2358 |
+
"eval_runtime": 32.8405,
|
| 2359 |
+
"eval_samples_per_second": 464.0,
|
| 2360 |
+
"eval_steps_per_second": 14.525,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 33.651659845384266,
|
| 2365 |
+
"grad_norm": 2.235302686691284,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.3415,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 33.651659845384266,
|
| 2372 |
+
"eval_accuracy": 0.9119416643480582,
|
| 2373 |
+
"eval_loss": 0.5329978466033936,
|
| 2374 |
+
"eval_runtime": 32.8284,
|
| 2375 |
+
"eval_samples_per_second": 464.171,
|
| 2376 |
+
"eval_steps_per_second": 14.53,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 33.879035925420645,
|
| 2381 |
+
"grad_norm": 2.2123124599456787,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.3471,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 33.879035925420645,
|
| 2388 |
+
"eval_accuracy": 0.9120227802759918,
|
| 2389 |
+
"eval_loss": 0.5121181607246399,
|
| 2390 |
+
"eval_runtime": 32.8709,
|
| 2391 |
+
"eval_samples_per_second": 463.571,
|
| 2392 |
+
"eval_steps_per_second": 14.511,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 34.106412005457024,
|
| 2397 |
+
"grad_norm": 2.3932435512542725,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.3458,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 34.106412005457024,
|
| 2404 |
+
"eval_accuracy": 0.9118813314037627,
|
| 2405 |
+
"eval_loss": 0.5218396186828613,
|
| 2406 |
+
"eval_runtime": 32.843,
|
| 2407 |
+
"eval_samples_per_second": 463.965,
|
| 2408 |
+
"eval_steps_per_second": 14.524,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 34.3337880854934,
|
| 2413 |
+
"grad_norm": 2.3429479598999023,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.3454,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 34.3337880854934,
|
| 2420 |
+
"eval_accuracy": 0.9123750564261301,
|
| 2421 |
+
"eval_loss": 0.5018845200538635,
|
| 2422 |
+
"eval_runtime": 33.1634,
|
| 2423 |
+
"eval_samples_per_second": 459.483,
|
| 2424 |
+
"eval_steps_per_second": 14.383,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 34.56116416552979,
|
| 2429 |
+
"grad_norm": 2.106468915939331,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.3437,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 34.56116416552979,
|
| 2436 |
+
"eval_accuracy": 0.9118771285864006,
|
| 2437 |
+
"eval_loss": 0.5132880210876465,
|
| 2438 |
+
"eval_runtime": 33.1566,
|
| 2439 |
+
"eval_samples_per_second": 459.577,
|
| 2440 |
+
"eval_steps_per_second": 14.386,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 34.78854024556617,
|
| 2445 |
+
"grad_norm": 2.7131636142730713,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.338,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 34.78854024556617,
|
| 2452 |
+
"eval_accuracy": 0.9125346722433042,
|
| 2453 |
+
"eval_loss": 0.5079774856567383,
|
| 2454 |
+
"eval_runtime": 32.8366,
|
| 2455 |
+
"eval_samples_per_second": 464.056,
|
| 2456 |
+
"eval_steps_per_second": 14.526,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 35.01591632560255,
|
| 2461 |
+
"grad_norm": 2.5895884037017822,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.337,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 35.01591632560255,
|
| 2468 |
+
"eval_accuracy": 0.9124522785288072,
|
| 2469 |
+
"eval_loss": 0.5153793692588806,
|
| 2470 |
+
"eval_runtime": 32.92,
|
| 2471 |
+
"eval_samples_per_second": 462.88,
|
| 2472 |
+
"eval_steps_per_second": 14.49,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 35.243292405638925,
|
| 2477 |
+
"grad_norm": 2.433104991912842,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.3315,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 35.243292405638925,
|
| 2484 |
+
"eval_accuracy": 0.9121799303487087,
|
| 2485 |
+
"eval_loss": 0.5197787284851074,
|
| 2486 |
+
"eval_runtime": 32.851,
|
| 2487 |
+
"eval_samples_per_second": 463.852,
|
| 2488 |
+
"eval_steps_per_second": 14.52,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 35.470668485675304,
|
| 2493 |
+
"grad_norm": 1.9214431047439575,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.3296,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 35.470668485675304,
|
| 2500 |
+
"eval_accuracy": 0.9126943544263368,
|
| 2501 |
+
"eval_loss": 0.5254490971565247,
|
| 2502 |
+
"eval_runtime": 32.8244,
|
| 2503 |
+
"eval_samples_per_second": 464.227,
|
| 2504 |
+
"eval_steps_per_second": 14.532,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 35.69804456571169,
|
| 2509 |
+
"grad_norm": 1.761439323425293,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.3368,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 35.69804456571169,
|
| 2516 |
+
"eval_accuracy": 0.9131551727198352,
|
| 2517 |
+
"eval_loss": 0.5223947167396545,
|
| 2518 |
+
"eval_runtime": 32.8821,
|
| 2519 |
+
"eval_samples_per_second": 463.413,
|
| 2520 |
+
"eval_steps_per_second": 14.506,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 35.92542064574807,
|
| 2525 |
+
"grad_norm": 2.2380154132843018,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.3327,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 35.92542064574807,
|
| 2532 |
+
"eval_accuracy": 0.9136349991769228,
|
| 2533 |
+
"eval_loss": 0.5105540156364441,
|
| 2534 |
+
"eval_runtime": 32.8762,
|
| 2535 |
+
"eval_samples_per_second": 463.496,
|
| 2536 |
+
"eval_steps_per_second": 14.509,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 36.15279672578445,
|
| 2541 |
+
"grad_norm": 3.250119209289551,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.3294,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 36.15279672578445,
|
| 2548 |
+
"eval_accuracy": 0.9137865895991087,
|
| 2549 |
+
"eval_loss": 0.5056498646736145,
|
| 2550 |
+
"eval_runtime": 32.8409,
|
| 2551 |
+
"eval_samples_per_second": 463.994,
|
| 2552 |
+
"eval_steps_per_second": 14.525,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 36.38017280582083,
|
| 2557 |
+
"grad_norm": 2.4665050506591797,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.3301,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 36.38017280582083,
|
| 2564 |
+
"eval_accuracy": 0.9142318719455823,
|
| 2565 |
+
"eval_loss": 0.5017902851104736,
|
| 2566 |
+
"eval_runtime": 32.7856,
|
| 2567 |
+
"eval_samples_per_second": 464.778,
|
| 2568 |
+
"eval_steps_per_second": 14.549,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 36.607548885857206,
|
| 2573 |
+
"grad_norm": 1.698752522468567,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.3225,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 36.607548885857206,
|
| 2580 |
+
"eval_accuracy": 0.9134595257800349,
|
| 2581 |
+
"eval_loss": 0.5074796080589294,
|
| 2582 |
+
"eval_runtime": 32.818,
|
| 2583 |
+
"eval_samples_per_second": 464.319,
|
| 2584 |
+
"eval_steps_per_second": 14.535,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 36.834924965893585,
|
| 2589 |
+
"grad_norm": 2.2286267280578613,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.3317,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 36.834924965893585,
|
| 2596 |
+
"eval_accuracy": 0.913460404731178,
|
| 2597 |
+
"eval_loss": 0.5129342675209045,
|
| 2598 |
+
"eval_runtime": 32.7859,
|
| 2599 |
+
"eval_samples_per_second": 464.772,
|
| 2600 |
+
"eval_steps_per_second": 14.549,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 37.06230104592997,
|
| 2605 |
+
"grad_norm": 2.5110349655151367,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.3268,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 37.06230104592997,
|
| 2612 |
+
"eval_accuracy": 0.9128155543200626,
|
| 2613 |
+
"eval_loss": 0.5117501020431519,
|
| 2614 |
+
"eval_runtime": 32.8322,
|
| 2615 |
+
"eval_samples_per_second": 464.118,
|
| 2616 |
+
"eval_steps_per_second": 14.528,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 37.28967712596635,
|
| 2621 |
+
"grad_norm": 2.6086220741271973,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.3232,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 37.28967712596635,
|
| 2628 |
+
"eval_accuracy": 0.9136776723576344,
|
| 2629 |
+
"eval_loss": 0.5069320797920227,
|
| 2630 |
+
"eval_runtime": 32.7874,
|
| 2631 |
+
"eval_samples_per_second": 464.751,
|
| 2632 |
+
"eval_steps_per_second": 14.548,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 37.51705320600273,
|
| 2637 |
+
"grad_norm": 2.095896005630493,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.3283,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 37.51705320600273,
|
| 2644 |
+
"eval_accuracy": 0.9137518563956867,
|
| 2645 |
+
"eval_loss": 0.5171787142753601,
|
| 2646 |
+
"eval_runtime": 31.7939,
|
| 2647 |
+
"eval_samples_per_second": 479.275,
|
| 2648 |
+
"eval_steps_per_second": 15.003,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 37.74442928603911,
|
| 2653 |
+
"grad_norm": 2.2822511196136475,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.3245,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 37.74442928603911,
|
| 2660 |
+
"eval_accuracy": 0.914121706237009,
|
| 2661 |
+
"eval_loss": 0.5052734613418579,
|
| 2662 |
+
"eval_runtime": 32.7523,
|
| 2663 |
+
"eval_samples_per_second": 465.25,
|
| 2664 |
+
"eval_steps_per_second": 14.564,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 37.971805366075486,
|
| 2669 |
+
"grad_norm": 2.3248562812805176,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.3212,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 37.971805366075486,
|
| 2676 |
+
"eval_accuracy": 0.9150295487683712,
|
| 2677 |
+
"eval_loss": 0.5122706890106201,
|
| 2678 |
+
"eval_runtime": 32.7649,
|
| 2679 |
+
"eval_samples_per_second": 465.071,
|
| 2680 |
+
"eval_steps_per_second": 14.558,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 38.19918144611187,
|
| 2685 |
+
"grad_norm": 2.381883382797241,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.3238,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 38.19918144611187,
|
| 2692 |
+
"eval_accuracy": 0.9140306949162653,
|
| 2693 |
+
"eval_loss": 0.5270175933837891,
|
| 2694 |
+
"eval_runtime": 32.7409,
|
| 2695 |
+
"eval_samples_per_second": 465.412,
|
| 2696 |
+
"eval_steps_per_second": 14.569,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 38.42655752614825,
|
| 2701 |
+
"grad_norm": 2.0925965309143066,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.3181,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 38.42655752614825,
|
| 2708 |
+
"eval_accuracy": 0.9146817787775174,
|
| 2709 |
+
"eval_loss": 0.5188115239143372,
|
| 2710 |
+
"eval_runtime": 32.7115,
|
| 2711 |
+
"eval_samples_per_second": 465.83,
|
| 2712 |
+
"eval_steps_per_second": 14.582,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 38.65393360618463,
|
| 2717 |
+
"grad_norm": 3.510465621948242,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.3235,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 38.65393360618463,
|
| 2724 |
+
"eval_accuracy": 0.9142990278989299,
|
| 2725 |
+
"eval_loss": 0.5162706971168518,
|
| 2726 |
+
"eval_runtime": 32.7411,
|
| 2727 |
+
"eval_samples_per_second": 465.409,
|
| 2728 |
+
"eval_steps_per_second": 14.569,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 38.88130968622101,
|
| 2733 |
+
"grad_norm": 1.8546830415725708,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.3202,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 38.88130968622101,
|
| 2740 |
+
"eval_accuracy": 0.9144765537609747,
|
| 2741 |
+
"eval_loss": 0.5259549021720886,
|
| 2742 |
+
"eval_runtime": 32.7768,
|
| 2743 |
+
"eval_samples_per_second": 464.903,
|
| 2744 |
+
"eval_steps_per_second": 14.553,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 39.10868576625739,
|
| 2749 |
+
"grad_norm": 2.1186509132385254,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.32,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 39.10868576625739,
|
| 2756 |
+
"eval_accuracy": 0.9149780487726317,
|
| 2757 |
+
"eval_loss": 0.5091920495033264,
|
| 2758 |
+
"eval_runtime": 32.7494,
|
| 2759 |
+
"eval_samples_per_second": 465.291,
|
| 2760 |
+
"eval_steps_per_second": 14.565,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 39.33606184629377,
|
| 2765 |
+
"grad_norm": 2.336721420288086,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.3187,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 39.33606184629377,
|
| 2772 |
+
"eval_accuracy": 0.9153005420364878,
|
| 2773 |
+
"eval_loss": 0.5176430344581604,
|
| 2774 |
+
"eval_runtime": 32.7347,
|
| 2775 |
+
"eval_samples_per_second": 465.5,
|
| 2776 |
+
"eval_steps_per_second": 14.572,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 39.56343792633015,
|
| 2781 |
+
"grad_norm": 2.6106879711151123,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.3091,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 39.56343792633015,
|
| 2788 |
+
"eval_accuracy": 0.9147028147603641,
|
| 2789 |
+
"eval_loss": 0.5056055784225464,
|
| 2790 |
+
"eval_runtime": 32.7536,
|
| 2791 |
+
"eval_samples_per_second": 465.231,
|
| 2792 |
+
"eval_steps_per_second": 14.563,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 39.79081400636653,
|
| 2797 |
+
"grad_norm": 1.6621290445327759,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.3226,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 39.79081400636653,
|
| 2804 |
+
"eval_accuracy": 0.9152877205965155,
|
| 2805 |
+
"eval_loss": 0.5065489411354065,
|
| 2806 |
+
"eval_runtime": 32.7739,
|
| 2807 |
+
"eval_samples_per_second": 464.943,
|
| 2808 |
+
"eval_steps_per_second": 14.554,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 40.01819008640291,
|
| 2813 |
+
"grad_norm": 2.2884457111358643,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.3154,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 40.01819008640291,
|
| 2820 |
+
"eval_accuracy": 0.9155226436727855,
|
| 2821 |
+
"eval_loss": 0.5103309750556946,
|
| 2822 |
+
"eval_runtime": 32.7346,
|
| 2823 |
+
"eval_samples_per_second": 465.502,
|
| 2824 |
+
"eval_steps_per_second": 14.572,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 40.24556616643929,
|
| 2829 |
+
"grad_norm": 2.1758711338043213,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.3136,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 40.24556616643929,
|
| 2836 |
+
"eval_accuracy": 0.9153148416356733,
|
| 2837 |
+
"eval_loss": 0.5262283682823181,
|
| 2838 |
+
"eval_runtime": 32.7358,
|
| 2839 |
+
"eval_samples_per_second": 465.485,
|
| 2840 |
+
"eval_steps_per_second": 14.571,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 40.47294224647567,
|
| 2845 |
+
"grad_norm": 2.209392547607422,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.3128,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 40.47294224647567,
|
| 2852 |
+
"eval_accuracy": 0.9157102933951855,
|
| 2853 |
+
"eval_loss": 0.5107696056365967,
|
| 2854 |
+
"eval_runtime": 32.7842,
|
| 2855 |
+
"eval_samples_per_second": 464.797,
|
| 2856 |
+
"eval_steps_per_second": 14.55,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 40.700318326512054,
|
| 2861 |
+
"grad_norm": 2.1529698371887207,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.3124,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 40.700318326512054,
|
| 2868 |
+
"eval_accuracy": 0.9159379589728691,
|
| 2869 |
+
"eval_loss": 0.5029435753822327,
|
| 2870 |
+
"eval_runtime": 32.7677,
|
| 2871 |
+
"eval_samples_per_second": 465.031,
|
| 2872 |
+
"eval_steps_per_second": 14.557,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 40.92769440654843,
|
| 2877 |
+
"grad_norm": 2.0045878887176514,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.3108,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 40.92769440654843,
|
| 2884 |
+
"eval_accuracy": 0.9157441455389975,
|
| 2885 |
+
"eval_loss": 0.5153588056564331,
|
| 2886 |
+
"eval_runtime": 32.7641,
|
| 2887 |
+
"eval_samples_per_second": 465.082,
|
| 2888 |
+
"eval_steps_per_second": 14.559,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 41.15507048658481,
|
| 2893 |
+
"grad_norm": 2.3361988067626953,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.3138,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 41.15507048658481,
|
| 2900 |
+
"eval_accuracy": 0.9159889763449391,
|
| 2901 |
+
"eval_loss": 0.5117628574371338,
|
| 2902 |
+
"eval_runtime": 32.7502,
|
| 2903 |
+
"eval_samples_per_second": 465.28,
|
| 2904 |
+
"eval_steps_per_second": 14.565,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 41.38244656662119,
|
| 2909 |
+
"grad_norm": 2.683246374130249,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.308,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 41.38244656662119,
|
| 2916 |
+
"eval_accuracy": 0.9162369011673905,
|
| 2917 |
+
"eval_loss": 0.49284523725509644,
|
| 2918 |
+
"eval_runtime": 32.7846,
|
| 2919 |
+
"eval_samples_per_second": 464.792,
|
| 2920 |
+
"eval_steps_per_second": 14.55,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 41.60982264665757,
|
| 2925 |
+
"grad_norm": 1.9992194175720215,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.3055,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 41.60982264665757,
|
| 2932 |
+
"eval_accuracy": 0.9158271965730912,
|
| 2933 |
+
"eval_loss": 0.511158287525177,
|
| 2934 |
+
"eval_runtime": 32.8203,
|
| 2935 |
+
"eval_samples_per_second": 464.286,
|
| 2936 |
+
"eval_steps_per_second": 14.534,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 41.83719872669395,
|
| 2941 |
+
"grad_norm": 1.5974903106689453,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.3093,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 41.83719872669395,
|
| 2948 |
+
"eval_accuracy": 0.915716300365121,
|
| 2949 |
+
"eval_loss": 0.512084424495697,
|
| 2950 |
+
"eval_runtime": 32.7946,
|
| 2951 |
+
"eval_samples_per_second": 464.649,
|
| 2952 |
+
"eval_steps_per_second": 14.545,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 42.064574806730334,
|
| 2957 |
+
"grad_norm": 1.9554269313812256,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.3074,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 42.064574806730334,
|
| 2964 |
+
"eval_accuracy": 0.9163171022735507,
|
| 2965 |
+
"eval_loss": 0.516119122505188,
|
| 2966 |
+
"eval_runtime": 32.7702,
|
| 2967 |
+
"eval_samples_per_second": 464.996,
|
| 2968 |
+
"eval_steps_per_second": 14.556,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 42.29195088676671,
|
| 2973 |
+
"grad_norm": 1.985527515411377,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.3017,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 42.29195088676671,
|
| 2980 |
+
"eval_accuracy": 0.917118710247863,
|
| 2981 |
+
"eval_loss": 0.5038302540779114,
|
| 2982 |
+
"eval_runtime": 32.8028,
|
| 2983 |
+
"eval_samples_per_second": 464.533,
|
| 2984 |
+
"eval_steps_per_second": 14.541,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 42.51932696680309,
|
| 2989 |
+
"grad_norm": 2.1202027797698975,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.3054,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 42.51932696680309,
|
| 2996 |
+
"eval_accuracy": 0.9161003507505029,
|
| 2997 |
+
"eval_loss": 0.5289775729179382,
|
| 2998 |
+
"eval_runtime": 32.7703,
|
| 2999 |
+
"eval_samples_per_second": 464.994,
|
| 3000 |
+
"eval_steps_per_second": 14.556,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 42.74670304683947,
|
| 3005 |
+
"grad_norm": 2.0218162536621094,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.3031,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 42.74670304683947,
|
| 3012 |
+
"eval_accuracy": 0.9163906164405633,
|
| 3013 |
+
"eval_loss": 0.5164464712142944,
|
| 3014 |
+
"eval_runtime": 32.7881,
|
| 3015 |
+
"eval_samples_per_second": 464.742,
|
| 3016 |
+
"eval_steps_per_second": 14.548,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 42.97407912687585,
|
| 3021 |
+
"grad_norm": 1.9983863830566406,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.3037,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 42.97407912687585,
|
| 3028 |
+
"eval_accuracy": 0.9169610509438041,
|
| 3029 |
+
"eval_loss": 0.5172046422958374,
|
| 3030 |
+
"eval_runtime": 32.7503,
|
| 3031 |
+
"eval_samples_per_second": 465.278,
|
| 3032 |
+
"eval_steps_per_second": 14.565,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 43.201455206912236,
|
| 3037 |
+
"grad_norm": 1.9231362342834473,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.3006,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 43.201455206912236,
|
| 3044 |
+
"eval_accuracy": 0.9164033951579469,
|
| 3045 |
+
"eval_loss": 0.5055718421936035,
|
| 3046 |
+
"eval_runtime": 32.8099,
|
| 3047 |
+
"eval_samples_per_second": 464.432,
|
| 3048 |
+
"eval_steps_per_second": 14.538,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 43.428831286948615,
|
| 3053 |
+
"grad_norm": 1.8182711601257324,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.2974,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 43.428831286948615,
|
| 3060 |
+
"eval_accuracy": 0.9165904140837446,
|
| 3061 |
+
"eval_loss": 0.5043260455131531,
|
| 3062 |
+
"eval_runtime": 32.7849,
|
| 3063 |
+
"eval_samples_per_second": 464.787,
|
| 3064 |
+
"eval_steps_per_second": 14.549,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 43.656207366984994,
|
| 3069 |
+
"grad_norm": 1.8277734518051147,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.3018,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 43.656207366984994,
|
| 3076 |
+
"eval_accuracy": 0.9161881058676558,
|
| 3077 |
+
"eval_loss": 0.49245980381965637,
|
| 3078 |
+
"eval_runtime": 32.7344,
|
| 3079 |
+
"eval_samples_per_second": 465.505,
|
| 3080 |
+
"eval_steps_per_second": 14.572,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 43.88358344702137,
|
| 3085 |
+
"grad_norm": 1.9253770112991333,
|
| 3086 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 3087 |
+
"loss": 0.3053,
|
| 3088 |
+
"step": 96500
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 43.88358344702137,
|
| 3092 |
+
"eval_accuracy": 0.9160392906798216,
|
| 3093 |
+
"eval_loss": 0.5248011350631714,
|
| 3094 |
+
"eval_runtime": 32.7872,
|
| 3095 |
+
"eval_samples_per_second": 464.754,
|
| 3096 |
+
"eval_steps_per_second": 14.548,
|
| 3097 |
+
"step": 96500
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 44.11095952705775,
|
| 3101 |
+
"grad_norm": 2.861922264099121,
|
| 3102 |
+
"learning_rate": 3e-06,
|
| 3103 |
+
"loss": 0.2994,
|
| 3104 |
+
"step": 97000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 44.11095952705775,
|
| 3108 |
+
"eval_accuracy": 0.9164925785172533,
|
| 3109 |
+
"eval_loss": 0.5156012177467346,
|
| 3110 |
+
"eval_runtime": 32.7667,
|
| 3111 |
+
"eval_samples_per_second": 465.045,
|
| 3112 |
+
"eval_steps_per_second": 14.557,
|
| 3113 |
+
"step": 97000
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 44.33833560709413,
|
| 3117 |
+
"grad_norm": 2.7760884761810303,
|
| 3118 |
+
"learning_rate": 2.5e-06,
|
| 3119 |
+
"loss": 0.3009,
|
| 3120 |
+
"step": 97500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 44.33833560709413,
|
| 3124 |
+
"eval_accuracy": 0.9166317350788379,
|
| 3125 |
+
"eval_loss": 0.5009229183197021,
|
| 3126 |
+
"eval_runtime": 32.7509,
|
| 3127 |
+
"eval_samples_per_second": 465.27,
|
| 3128 |
+
"eval_steps_per_second": 14.565,
|
| 3129 |
+
"step": 97500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 44.565711687130516,
|
| 3133 |
+
"grad_norm": 2.3244969844818115,
|
| 3134 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 3135 |
+
"loss": 0.301,
|
| 3136 |
+
"step": 98000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 44.565711687130516,
|
| 3140 |
+
"eval_accuracy": 0.9171028513895522,
|
| 3141 |
+
"eval_loss": 0.5136500000953674,
|
| 3142 |
+
"eval_runtime": 32.7529,
|
| 3143 |
+
"eval_samples_per_second": 465.242,
|
| 3144 |
+
"eval_steps_per_second": 14.564,
|
| 3145 |
+
"step": 98000
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 44.793087767166895,
|
| 3149 |
+
"grad_norm": 1.7618470191955566,
|
| 3150 |
+
"learning_rate": 1.5e-06,
|
| 3151 |
+
"loss": 0.3014,
|
| 3152 |
+
"step": 98500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 44.793087767166895,
|
| 3156 |
+
"eval_accuracy": 0.9166436355819753,
|
| 3157 |
+
"eval_loss": 0.5133616924285889,
|
| 3158 |
+
"eval_runtime": 32.7772,
|
| 3159 |
+
"eval_samples_per_second": 464.896,
|
| 3160 |
+
"eval_steps_per_second": 14.553,
|
| 3161 |
+
"step": 98500
|
| 3162 |
+
},
|
| 3163 |
+
{
|
| 3164 |
+
"epoch": 45.020463847203274,
|
| 3165 |
+
"grad_norm": 2.309792995452881,
|
| 3166 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 3167 |
+
"loss": 0.2971,
|
| 3168 |
+
"step": 99000
|
| 3169 |
+
},
|
| 3170 |
+
{
|
| 3171 |
+
"epoch": 45.020463847203274,
|
| 3172 |
+
"eval_accuracy": 0.9174242277399804,
|
| 3173 |
+
"eval_loss": 0.4882655441761017,
|
| 3174 |
+
"eval_runtime": 32.7744,
|
| 3175 |
+
"eval_samples_per_second": 464.936,
|
| 3176 |
+
"eval_steps_per_second": 14.554,
|
| 3177 |
+
"step": 99000
|
| 3178 |
+
}
|
| 3179 |
+
],
|
| 3180 |
+
"logging_steps": 500,
|
| 3181 |
+
"max_steps": 100000,
|
| 3182 |
+
"num_input_tokens_seen": 0,
|
| 3183 |
+
"num_train_epochs": 46,
|
| 3184 |
+
"save_steps": 500,
|
| 3185 |
+
"stateful_callbacks": {
|
| 3186 |
+
"TrainerControl": {
|
| 3187 |
+
"args": {
|
| 3188 |
+
"should_epoch_stop": false,
|
| 3189 |
+
"should_evaluate": false,
|
| 3190 |
+
"should_log": false,
|
| 3191 |
+
"should_save": true,
|
| 3192 |
+
"should_training_stop": false
|
| 3193 |
+
},
|
| 3194 |
+
"attributes": {}
|
| 3195 |
+
}
|
| 3196 |
+
},
|
| 3197 |
+
"total_flos": 8.346580545102152e+17,
|
| 3198 |
+
"train_batch_size": 16,
|
| 3199 |
+
"trial_name": null,
|
| 3200 |
+
"trial_params": null
|
| 3201 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f4d6cbb77dd621f455f5e9a0355b01b8c8d2729f126f54a9bf8cafe1362f509
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|