Uploading checkpoint-98500 for mbert - uzn-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 +3185 -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: uzn-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 |
+
# uzn-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: 0.4191
|
| 22 |
+
- Accuracy: 0.9085
|
| 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 sha256:27757e9145101eeb4eed501e01cd00f90057cbf066b8ed052d3c6b7942bac427
|
| 3 |
+
size 711921812
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16cd77838594f2d9b9a6c13e52ac3de766d1b12dc33e5e5f9e7fa1880efaff5f
|
| 3 |
+
size 1423965114
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd98828ef391a9bd7d7f37e992ed5cd1e01071bff64e614960e47dc7c38f3f1a
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54e4dc8cb5bda128de11b31af58dc4443a5faadde8b602fc3ef875e446d3ac98
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b72c24b8eeac82a0d428f6aaf2686dbda50deecaee6ca497e01bfbb1a32f05c
|
| 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 +1,3185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4168604016304016,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/mbert/uzn-Latn/checkpoint-98500",
|
| 4 |
+
"epoch": 41.684299619128225,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 98500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21159542953872196,
|
| 13 |
+
"grad_norm": 2.921497344970703,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.4276,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.21159542953872196,
|
| 20 |
+
"eval_accuracy": 0.7739727645766222,
|
| 21 |
+
"eval_loss": 1.1149295568466187,
|
| 22 |
+
"eval_runtime": 56.3131,
|
| 23 |
+
"eval_samples_per_second": 335.57,
|
| 24 |
+
"eval_steps_per_second": 10.495,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.4231908590774439,
|
| 29 |
+
"grad_norm": 2.887310743331909,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.1534,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.4231908590774439,
|
| 36 |
+
"eval_accuracy": 0.796249290534149,
|
| 37 |
+
"eval_loss": 1.0010449886322021,
|
| 38 |
+
"eval_runtime": 56.6477,
|
| 39 |
+
"eval_samples_per_second": 333.588,
|
| 40 |
+
"eval_steps_per_second": 10.433,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.6347862886161659,
|
| 45 |
+
"grad_norm": 2.78161883354187,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.052,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.6347862886161659,
|
| 52 |
+
"eval_accuracy": 0.8080684952757127,
|
| 53 |
+
"eval_loss": 0.9241422414779663,
|
| 54 |
+
"eval_runtime": 56.6749,
|
| 55 |
+
"eval_samples_per_second": 333.428,
|
| 56 |
+
"eval_steps_per_second": 10.428,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.8463817181548878,
|
| 61 |
+
"grad_norm": 3.055419921875,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 0.9862,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.8463817181548878,
|
| 68 |
+
"eval_accuracy": 0.8152558855004783,
|
| 69 |
+
"eval_loss": 0.8904742002487183,
|
| 70 |
+
"eval_runtime": 56.6659,
|
| 71 |
+
"eval_samples_per_second": 333.481,
|
| 72 |
+
"eval_steps_per_second": 10.43,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.0579771476936097,
|
| 77 |
+
"grad_norm": 2.547567367553711,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 0.9386,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0579771476936097,
|
| 84 |
+
"eval_accuracy": 0.8231659750971954,
|
| 85 |
+
"eval_loss": 0.8489812016487122,
|
| 86 |
+
"eval_runtime": 56.6994,
|
| 87 |
+
"eval_samples_per_second": 333.284,
|
| 88 |
+
"eval_steps_per_second": 10.423,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.2695725772323319,
|
| 93 |
+
"grad_norm": 2.394988536834717,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 0.9015,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.2695725772323319,
|
| 100 |
+
"eval_accuracy": 0.828950558445466,
|
| 101 |
+
"eval_loss": 0.8159027695655823,
|
| 102 |
+
"eval_runtime": 56.819,
|
| 103 |
+
"eval_samples_per_second": 332.582,
|
| 104 |
+
"eval_steps_per_second": 10.401,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.4811680067710538,
|
| 109 |
+
"grad_norm": 2.5777626037597656,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 0.8766,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.4811680067710538,
|
| 116 |
+
"eval_accuracy": 0.8325975068653477,
|
| 117 |
+
"eval_loss": 0.8011560440063477,
|
| 118 |
+
"eval_runtime": 56.6442,
|
| 119 |
+
"eval_samples_per_second": 333.609,
|
| 120 |
+
"eval_steps_per_second": 10.434,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.6927634363097757,
|
| 125 |
+
"grad_norm": 2.2821884155273438,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 0.8577,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.6927634363097757,
|
| 132 |
+
"eval_accuracy": 0.8376275083743905,
|
| 133 |
+
"eval_loss": 0.7690746784210205,
|
| 134 |
+
"eval_runtime": 56.6871,
|
| 135 |
+
"eval_samples_per_second": 333.356,
|
| 136 |
+
"eval_steps_per_second": 10.426,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.9043588658484976,
|
| 141 |
+
"grad_norm": 2.3086235523223877,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 0.8333,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.9043588658484976,
|
| 148 |
+
"eval_accuracy": 0.8403614576449804,
|
| 149 |
+
"eval_loss": 0.7538630366325378,
|
| 150 |
+
"eval_runtime": 56.6299,
|
| 151 |
+
"eval_samples_per_second": 333.693,
|
| 152 |
+
"eval_steps_per_second": 10.436,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.1159542953872195,
|
| 157 |
+
"grad_norm": 2.47580885887146,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 0.8086,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.1159542953872195,
|
| 164 |
+
"eval_accuracy": 0.8421322737662698,
|
| 165 |
+
"eval_loss": 0.7495782375335693,
|
| 166 |
+
"eval_runtime": 56.7249,
|
| 167 |
+
"eval_samples_per_second": 333.134,
|
| 168 |
+
"eval_steps_per_second": 10.419,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.327549724925942,
|
| 173 |
+
"grad_norm": 2.1193575859069824,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 0.7915,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.327549724925942,
|
| 180 |
+
"eval_accuracy": 0.8444147387701536,
|
| 181 |
+
"eval_loss": 0.7350887656211853,
|
| 182 |
+
"eval_runtime": 56.6712,
|
| 183 |
+
"eval_samples_per_second": 333.45,
|
| 184 |
+
"eval_steps_per_second": 10.429,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.5391451544646637,
|
| 189 |
+
"grad_norm": 2.541102409362793,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 0.7833,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.5391451544646637,
|
| 196 |
+
"eval_accuracy": 0.8466142754406873,
|
| 197 |
+
"eval_loss": 0.7291632294654846,
|
| 198 |
+
"eval_runtime": 56.6678,
|
| 199 |
+
"eval_samples_per_second": 333.47,
|
| 200 |
+
"eval_steps_per_second": 10.429,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.7507405840033856,
|
| 205 |
+
"grad_norm": 2.231060028076172,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 0.772,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.7507405840033856,
|
| 212 |
+
"eval_accuracy": 0.8490486197954219,
|
| 213 |
+
"eval_loss": 0.711447536945343,
|
| 214 |
+
"eval_runtime": 56.6668,
|
| 215 |
+
"eval_samples_per_second": 333.476,
|
| 216 |
+
"eval_steps_per_second": 10.429,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.9623360135421075,
|
| 221 |
+
"grad_norm": 2.1715972423553467,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.7581,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.9623360135421075,
|
| 228 |
+
"eval_accuracy": 0.8519245359125641,
|
| 229 |
+
"eval_loss": 0.6959145069122314,
|
| 230 |
+
"eval_runtime": 56.6787,
|
| 231 |
+
"eval_samples_per_second": 333.406,
|
| 232 |
+
"eval_steps_per_second": 10.427,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.1739314430808294,
|
| 237 |
+
"grad_norm": 2.2937397956848145,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.7415,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.1739314430808294,
|
| 244 |
+
"eval_accuracy": 0.8529956231046303,
|
| 245 |
+
"eval_loss": 0.6926664113998413,
|
| 246 |
+
"eval_runtime": 56.6521,
|
| 247 |
+
"eval_samples_per_second": 333.562,
|
| 248 |
+
"eval_steps_per_second": 10.432,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.3855268726195513,
|
| 253 |
+
"grad_norm": 2.5523250102996826,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.736,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.3855268726195513,
|
| 260 |
+
"eval_accuracy": 0.8532365311483308,
|
| 261 |
+
"eval_loss": 0.6907092332839966,
|
| 262 |
+
"eval_runtime": 56.7139,
|
| 263 |
+
"eval_samples_per_second": 333.199,
|
| 264 |
+
"eval_steps_per_second": 10.421,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 3.597122302158273,
|
| 269 |
+
"grad_norm": 2.0444934368133545,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.7296,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.597122302158273,
|
| 276 |
+
"eval_accuracy": 0.8552509198450057,
|
| 277 |
+
"eval_loss": 0.6779365539550781,
|
| 278 |
+
"eval_runtime": 56.7215,
|
| 279 |
+
"eval_samples_per_second": 333.154,
|
| 280 |
+
"eval_steps_per_second": 10.419,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.808717731696995,
|
| 285 |
+
"grad_norm": 2.0448098182678223,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.7205,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.808717731696995,
|
| 292 |
+
"eval_accuracy": 0.8566500457200734,
|
| 293 |
+
"eval_loss": 0.6696876883506775,
|
| 294 |
+
"eval_runtime": 56.6572,
|
| 295 |
+
"eval_samples_per_second": 333.532,
|
| 296 |
+
"eval_steps_per_second": 10.431,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.020313161235717,
|
| 301 |
+
"grad_norm": 2.0469236373901367,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.7188,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.020313161235717,
|
| 308 |
+
"eval_accuracy": 0.8580638342011417,
|
| 309 |
+
"eval_loss": 0.6623690724372864,
|
| 310 |
+
"eval_runtime": 56.6848,
|
| 311 |
+
"eval_samples_per_second": 333.37,
|
| 312 |
+
"eval_steps_per_second": 10.426,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 4.231908590774439,
|
| 317 |
+
"grad_norm": 2.1119368076324463,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.6982,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.231908590774439,
|
| 324 |
+
"eval_accuracy": 0.8593053877130069,
|
| 325 |
+
"eval_loss": 0.659415602684021,
|
| 326 |
+
"eval_runtime": 56.6748,
|
| 327 |
+
"eval_samples_per_second": 333.429,
|
| 328 |
+
"eval_steps_per_second": 10.428,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 4.443504020313161,
|
| 333 |
+
"grad_norm": 2.189457654953003,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.6959,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.443504020313161,
|
| 340 |
+
"eval_accuracy": 0.8603594080338266,
|
| 341 |
+
"eval_loss": 0.657759428024292,
|
| 342 |
+
"eval_runtime": 56.6831,
|
| 343 |
+
"eval_samples_per_second": 333.38,
|
| 344 |
+
"eval_steps_per_second": 10.426,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.655099449851884,
|
| 349 |
+
"grad_norm": 1.929677963256836,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.6929,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 4.655099449851884,
|
| 356 |
+
"eval_accuracy": 0.8612637514250079,
|
| 357 |
+
"eval_loss": 0.6502240300178528,
|
| 358 |
+
"eval_runtime": 56.7024,
|
| 359 |
+
"eval_samples_per_second": 333.266,
|
| 360 |
+
"eval_steps_per_second": 10.423,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 4.8666948793906055,
|
| 365 |
+
"grad_norm": 1.8653119802474976,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.6822,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.8666948793906055,
|
| 372 |
+
"eval_accuracy": 0.8619526948014083,
|
| 373 |
+
"eval_loss": 0.6439126133918762,
|
| 374 |
+
"eval_runtime": 56.6628,
|
| 375 |
+
"eval_samples_per_second": 333.499,
|
| 376 |
+
"eval_steps_per_second": 10.43,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 5.078290308929327,
|
| 381 |
+
"grad_norm": 2.043660879135132,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.6791,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.078290308929327,
|
| 388 |
+
"eval_accuracy": 0.8633964837614313,
|
| 389 |
+
"eval_loss": 0.6387478709220886,
|
| 390 |
+
"eval_runtime": 56.6638,
|
| 391 |
+
"eval_samples_per_second": 333.493,
|
| 392 |
+
"eval_steps_per_second": 10.43,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 5.289885738468049,
|
| 397 |
+
"grad_norm": 2.096825361251831,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.6651,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 5.289885738468049,
|
| 404 |
+
"eval_accuracy": 0.8642863103795229,
|
| 405 |
+
"eval_loss": 0.6335816383361816,
|
| 406 |
+
"eval_runtime": 56.6782,
|
| 407 |
+
"eval_samples_per_second": 333.408,
|
| 408 |
+
"eval_steps_per_second": 10.427,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 5.501481168006771,
|
| 413 |
+
"grad_norm": 1.928902268409729,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.6619,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.501481168006771,
|
| 420 |
+
"eval_accuracy": 0.8654069776602684,
|
| 421 |
+
"eval_loss": 0.6258506774902344,
|
| 422 |
+
"eval_runtime": 56.6693,
|
| 423 |
+
"eval_samples_per_second": 333.461,
|
| 424 |
+
"eval_steps_per_second": 10.429,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.713076597545493,
|
| 429 |
+
"grad_norm": 2.0632073879241943,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.6682,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 5.713076597545493,
|
| 436 |
+
"eval_accuracy": 0.8657809739996617,
|
| 437 |
+
"eval_loss": 0.6195570230484009,
|
| 438 |
+
"eval_runtime": 56.731,
|
| 439 |
+
"eval_samples_per_second": 333.098,
|
| 440 |
+
"eval_steps_per_second": 10.418,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.924672027084215,
|
| 445 |
+
"grad_norm": 1.9731072187423706,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.6565,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.924672027084215,
|
| 452 |
+
"eval_accuracy": 0.8663846700592414,
|
| 453 |
+
"eval_loss": 0.6176990866661072,
|
| 454 |
+
"eval_runtime": 56.6314,
|
| 455 |
+
"eval_samples_per_second": 333.684,
|
| 456 |
+
"eval_steps_per_second": 10.436,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 6.136267456622937,
|
| 461 |
+
"grad_norm": 1.908463954925537,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.6435,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.136267456622937,
|
| 468 |
+
"eval_accuracy": 0.8674512611356094,
|
| 469 |
+
"eval_loss": 0.6195366978645325,
|
| 470 |
+
"eval_runtime": 56.655,
|
| 471 |
+
"eval_samples_per_second": 333.545,
|
| 472 |
+
"eval_steps_per_second": 10.432,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 6.347862886161659,
|
| 477 |
+
"grad_norm": 1.9294794797897339,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.6473,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 6.347862886161659,
|
| 484 |
+
"eval_accuracy": 0.8688036504176064,
|
| 485 |
+
"eval_loss": 0.6084388494491577,
|
| 486 |
+
"eval_runtime": 56.7044,
|
| 487 |
+
"eval_samples_per_second": 333.255,
|
| 488 |
+
"eval_steps_per_second": 10.422,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 6.559458315700381,
|
| 493 |
+
"grad_norm": 1.822090983390808,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.6405,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 6.559458315700381,
|
| 500 |
+
"eval_accuracy": 0.8688105679087649,
|
| 501 |
+
"eval_loss": 0.6102983951568604,
|
| 502 |
+
"eval_runtime": 56.6909,
|
| 503 |
+
"eval_samples_per_second": 333.334,
|
| 504 |
+
"eval_steps_per_second": 10.425,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 6.771053745239103,
|
| 509 |
+
"grad_norm": 1.8274174928665161,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.6394,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 6.771053745239103,
|
| 516 |
+
"eval_accuracy": 0.8696247275715319,
|
| 517 |
+
"eval_loss": 0.6044107675552368,
|
| 518 |
+
"eval_runtime": 56.6909,
|
| 519 |
+
"eval_samples_per_second": 333.334,
|
| 520 |
+
"eval_steps_per_second": 10.425,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 6.9826491747778245,
|
| 525 |
+
"grad_norm": 1.957808256149292,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.6323,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 6.9826491747778245,
|
| 532 |
+
"eval_accuracy": 0.8709820384830048,
|
| 533 |
+
"eval_loss": 0.6035895347595215,
|
| 534 |
+
"eval_runtime": 56.7516,
|
| 535 |
+
"eval_samples_per_second": 332.978,
|
| 536 |
+
"eval_steps_per_second": 10.414,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 7.194244604316546,
|
| 541 |
+
"grad_norm": 2.1266331672668457,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.6242,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 7.194244604316546,
|
| 548 |
+
"eval_accuracy": 0.8710693061823496,
|
| 549 |
+
"eval_loss": 0.5997599363327026,
|
| 550 |
+
"eval_runtime": 56.6612,
|
| 551 |
+
"eval_samples_per_second": 333.509,
|
| 552 |
+
"eval_steps_per_second": 10.43,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 7.405840033855268,
|
| 557 |
+
"grad_norm": 2.0006179809570312,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.6203,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 7.405840033855268,
|
| 564 |
+
"eval_accuracy": 0.871826703965482,
|
| 565 |
+
"eval_loss": 0.593313455581665,
|
| 566 |
+
"eval_runtime": 56.7083,
|
| 567 |
+
"eval_samples_per_second": 333.232,
|
| 568 |
+
"eval_steps_per_second": 10.422,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 7.61743546339399,
|
| 573 |
+
"grad_norm": 1.9261244535446167,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.6188,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 7.61743546339399,
|
| 580 |
+
"eval_accuracy": 0.8720450520950311,
|
| 581 |
+
"eval_loss": 0.5938737392425537,
|
| 582 |
+
"eval_runtime": 56.6704,
|
| 583 |
+
"eval_samples_per_second": 333.454,
|
| 584 |
+
"eval_steps_per_second": 10.429,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 7.829030892932713,
|
| 589 |
+
"grad_norm": 2.1938488483428955,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.6208,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 7.829030892932713,
|
| 596 |
+
"eval_accuracy": 0.8725229150021704,
|
| 597 |
+
"eval_loss": 0.5889946222305298,
|
| 598 |
+
"eval_runtime": 56.7137,
|
| 599 |
+
"eval_samples_per_second": 333.2,
|
| 600 |
+
"eval_steps_per_second": 10.421,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 8.040626322471434,
|
| 605 |
+
"grad_norm": 2.1036553382873535,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.613,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 8.040626322471434,
|
| 612 |
+
"eval_accuracy": 0.8735021854184964,
|
| 613 |
+
"eval_loss": 0.588358461856842,
|
| 614 |
+
"eval_runtime": 56.6871,
|
| 615 |
+
"eval_samples_per_second": 333.356,
|
| 616 |
+
"eval_steps_per_second": 10.426,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 8.252221752010156,
|
| 621 |
+
"grad_norm": 1.9046729803085327,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.603,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 8.252221752010156,
|
| 628 |
+
"eval_accuracy": 0.8738945036710153,
|
| 629 |
+
"eval_loss": 0.5827779769897461,
|
| 630 |
+
"eval_runtime": 56.6965,
|
| 631 |
+
"eval_samples_per_second": 333.301,
|
| 632 |
+
"eval_steps_per_second": 10.424,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 8.463817181548878,
|
| 637 |
+
"grad_norm": 1.9339730739593506,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.5994,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 8.463817181548878,
|
| 644 |
+
"eval_accuracy": 0.875056937375979,
|
| 645 |
+
"eval_loss": 0.5773849487304688,
|
| 646 |
+
"eval_runtime": 56.7994,
|
| 647 |
+
"eval_samples_per_second": 332.697,
|
| 648 |
+
"eval_steps_per_second": 10.405,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 8.6754126110876,
|
| 653 |
+
"grad_norm": 2.686779737472534,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.6069,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 8.6754126110876,
|
| 660 |
+
"eval_accuracy": 0.8750560557996572,
|
| 661 |
+
"eval_loss": 0.5813719630241394,
|
| 662 |
+
"eval_runtime": 56.6685,
|
| 663 |
+
"eval_samples_per_second": 333.465,
|
| 664 |
+
"eval_steps_per_second": 10.429,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 8.887008040626322,
|
| 669 |
+
"grad_norm": 1.8793710470199585,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.5978,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 8.887008040626322,
|
| 676 |
+
"eval_accuracy": 0.8751772357288334,
|
| 677 |
+
"eval_loss": 0.573668360710144,
|
| 678 |
+
"eval_runtime": 56.7394,
|
| 679 |
+
"eval_samples_per_second": 333.049,
|
| 680 |
+
"eval_steps_per_second": 10.416,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 9.098603470165044,
|
| 685 |
+
"grad_norm": 2.0106098651885986,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.5949,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 9.098603470165044,
|
| 692 |
+
"eval_accuracy": 0.8763350356219658,
|
| 693 |
+
"eval_loss": 0.5748294591903687,
|
| 694 |
+
"eval_runtime": 56.7185,
|
| 695 |
+
"eval_samples_per_second": 333.172,
|
| 696 |
+
"eval_steps_per_second": 10.42,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 9.310198899703767,
|
| 701 |
+
"grad_norm": 2.0075221061706543,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.5882,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 9.310198899703767,
|
| 708 |
+
"eval_accuracy": 0.8762109209507963,
|
| 709 |
+
"eval_loss": 0.5719663500785828,
|
| 710 |
+
"eval_runtime": 56.7186,
|
| 711 |
+
"eval_samples_per_second": 333.171,
|
| 712 |
+
"eval_steps_per_second": 10.42,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 9.52179432924249,
|
| 717 |
+
"grad_norm": 2.142890453338623,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.5851,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 9.52179432924249,
|
| 724 |
+
"eval_accuracy": 0.8763193666724947,
|
| 725 |
+
"eval_loss": 0.5695766806602478,
|
| 726 |
+
"eval_runtime": 56.7021,
|
| 727 |
+
"eval_samples_per_second": 333.268,
|
| 728 |
+
"eval_steps_per_second": 10.423,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 9.733389758781211,
|
| 733 |
+
"grad_norm": 1.9425110816955566,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.5919,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 9.733389758781211,
|
| 740 |
+
"eval_accuracy": 0.8773136748614612,
|
| 741 |
+
"eval_loss": 0.5677768588066101,
|
| 742 |
+
"eval_runtime": 56.7063,
|
| 743 |
+
"eval_samples_per_second": 333.243,
|
| 744 |
+
"eval_steps_per_second": 10.422,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 9.944985188319933,
|
| 749 |
+
"grad_norm": 1.9026435613632202,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.5816,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 9.944985188319933,
|
| 756 |
+
"eval_accuracy": 0.8777507496950493,
|
| 757 |
+
"eval_loss": 0.5624855160713196,
|
| 758 |
+
"eval_runtime": 56.7191,
|
| 759 |
+
"eval_samples_per_second": 333.168,
|
| 760 |
+
"eval_steps_per_second": 10.42,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 10.156580617858655,
|
| 765 |
+
"grad_norm": 1.9598579406738281,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.5756,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 10.156580617858655,
|
| 772 |
+
"eval_accuracy": 0.8792867643467259,
|
| 773 |
+
"eval_loss": 0.5589098930358887,
|
| 774 |
+
"eval_runtime": 56.7351,
|
| 775 |
+
"eval_samples_per_second": 333.074,
|
| 776 |
+
"eval_steps_per_second": 10.417,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 10.368176047397377,
|
| 781 |
+
"grad_norm": 1.8479692935943604,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.5748,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 10.368176047397377,
|
| 788 |
+
"eval_accuracy": 0.8791260366030839,
|
| 789 |
+
"eval_loss": 0.5566587448120117,
|
| 790 |
+
"eval_runtime": 56.6949,
|
| 791 |
+
"eval_samples_per_second": 333.31,
|
| 792 |
+
"eval_steps_per_second": 10.424,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 10.579771476936099,
|
| 797 |
+
"grad_norm": 1.5431069135665894,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.58,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 10.579771476936099,
|
| 804 |
+
"eval_accuracy": 0.8793816874556923,
|
| 805 |
+
"eval_loss": 0.5564578175544739,
|
| 806 |
+
"eval_runtime": 56.6656,
|
| 807 |
+
"eval_samples_per_second": 333.483,
|
| 808 |
+
"eval_steps_per_second": 10.43,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 10.79136690647482,
|
| 813 |
+
"grad_norm": 1.6908243894577026,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.5697,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 10.79136690647482,
|
| 820 |
+
"eval_accuracy": 0.8798802132166955,
|
| 821 |
+
"eval_loss": 0.5517674684524536,
|
| 822 |
+
"eval_runtime": 56.7133,
|
| 823 |
+
"eval_samples_per_second": 333.202,
|
| 824 |
+
"eval_steps_per_second": 10.421,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 11.002962336013542,
|
| 829 |
+
"grad_norm": 1.613221287727356,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.5677,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 11.002962336013542,
|
| 836 |
+
"eval_accuracy": 0.8803581109191997,
|
| 837 |
+
"eval_loss": 0.5512924790382385,
|
| 838 |
+
"eval_runtime": 56.6592,
|
| 839 |
+
"eval_samples_per_second": 333.521,
|
| 840 |
+
"eval_steps_per_second": 10.431,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 11.214557765552264,
|
| 845 |
+
"grad_norm": 1.9331544637680054,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.5638,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 11.214557765552264,
|
| 852 |
+
"eval_accuracy": 0.8803611347658262,
|
| 853 |
+
"eval_loss": 0.5520021319389343,
|
| 854 |
+
"eval_runtime": 56.7111,
|
| 855 |
+
"eval_samples_per_second": 333.215,
|
| 856 |
+
"eval_steps_per_second": 10.421,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 11.426153195090986,
|
| 861 |
+
"grad_norm": 1.776992678642273,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.5559,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 11.426153195090986,
|
| 868 |
+
"eval_accuracy": 0.8818474868190084,
|
| 869 |
+
"eval_loss": 0.5454077124595642,
|
| 870 |
+
"eval_runtime": 56.6928,
|
| 871 |
+
"eval_samples_per_second": 333.323,
|
| 872 |
+
"eval_steps_per_second": 10.425,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 11.637748624629708,
|
| 877 |
+
"grad_norm": 1.8591481447219849,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.5588,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 11.637748624629708,
|
| 884 |
+
"eval_accuracy": 0.8813020083937573,
|
| 885 |
+
"eval_loss": 0.5480958223342896,
|
| 886 |
+
"eval_runtime": 56.6926,
|
| 887 |
+
"eval_samples_per_second": 333.324,
|
| 888 |
+
"eval_steps_per_second": 10.425,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 11.84934405416843,
|
| 893 |
+
"grad_norm": 2.0454280376434326,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.5601,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 11.84934405416843,
|
| 900 |
+
"eval_accuracy": 0.8815235988618855,
|
| 901 |
+
"eval_loss": 0.5503857731819153,
|
| 902 |
+
"eval_runtime": 56.6829,
|
| 903 |
+
"eval_samples_per_second": 333.381,
|
| 904 |
+
"eval_steps_per_second": 10.426,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 12.060939483707152,
|
| 909 |
+
"grad_norm": 1.7857919931411743,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.5602,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 12.060939483707152,
|
| 916 |
+
"eval_accuracy": 0.8822896243277506,
|
| 917 |
+
"eval_loss": 0.5415288805961609,
|
| 918 |
+
"eval_runtime": 56.7189,
|
| 919 |
+
"eval_samples_per_second": 333.169,
|
| 920 |
+
"eval_steps_per_second": 10.42,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 12.272534913245874,
|
| 925 |
+
"grad_norm": 1.8363231420516968,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.5495,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 12.272534913245874,
|
| 932 |
+
"eval_accuracy": 0.8819665084878122,
|
| 933 |
+
"eval_loss": 0.5420069098472595,
|
| 934 |
+
"eval_runtime": 56.7641,
|
| 935 |
+
"eval_samples_per_second": 332.904,
|
| 936 |
+
"eval_steps_per_second": 10.412,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 12.484130342784596,
|
| 941 |
+
"grad_norm": 1.8736658096313477,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.5504,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 12.484130342784596,
|
| 948 |
+
"eval_accuracy": 0.8835764719980574,
|
| 949 |
+
"eval_loss": 0.5378038287162781,
|
| 950 |
+
"eval_runtime": 56.7267,
|
| 951 |
+
"eval_samples_per_second": 333.123,
|
| 952 |
+
"eval_steps_per_second": 10.418,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 12.695725772323318,
|
| 957 |
+
"grad_norm": 1.8123923540115356,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.5437,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 12.695725772323318,
|
| 964 |
+
"eval_accuracy": 0.8830367484137376,
|
| 965 |
+
"eval_loss": 0.5415025949478149,
|
| 966 |
+
"eval_runtime": 56.7723,
|
| 967 |
+
"eval_samples_per_second": 332.856,
|
| 968 |
+
"eval_steps_per_second": 10.41,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 12.90732120186204,
|
| 973 |
+
"grad_norm": 1.7283321619033813,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.547,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 12.90732120186204,
|
| 980 |
+
"eval_accuracy": 0.8831898301210944,
|
| 981 |
+
"eval_loss": 0.5370810627937317,
|
| 982 |
+
"eval_runtime": 56.7048,
|
| 983 |
+
"eval_samples_per_second": 333.252,
|
| 984 |
+
"eval_steps_per_second": 10.422,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 13.118916631400761,
|
| 989 |
+
"grad_norm": 1.7546370029449463,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.5421,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 13.118916631400761,
|
| 996 |
+
"eval_accuracy": 0.8843731246662792,
|
| 997 |
+
"eval_loss": 0.5339444875717163,
|
| 998 |
+
"eval_runtime": 56.7512,
|
| 999 |
+
"eval_samples_per_second": 332.98,
|
| 1000 |
+
"eval_steps_per_second": 10.414,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 13.330512060939483,
|
| 1005 |
+
"grad_norm": 1.7105295658111572,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.5292,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 13.330512060939483,
|
| 1012 |
+
"eval_accuracy": 0.8839966938446517,
|
| 1013 |
+
"eval_loss": 0.5369998216629028,
|
| 1014 |
+
"eval_runtime": 56.7122,
|
| 1015 |
+
"eval_samples_per_second": 333.209,
|
| 1016 |
+
"eval_steps_per_second": 10.421,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 13.542107490478205,
|
| 1021 |
+
"grad_norm": 1.981752872467041,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.5377,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 13.542107490478205,
|
| 1028 |
+
"eval_accuracy": 0.8838718555622833,
|
| 1029 |
+
"eval_loss": 0.5359402298927307,
|
| 1030 |
+
"eval_runtime": 56.7175,
|
| 1031 |
+
"eval_samples_per_second": 333.178,
|
| 1032 |
+
"eval_steps_per_second": 10.42,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 13.753702920016927,
|
| 1037 |
+
"grad_norm": 1.7901474237442017,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.5368,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 13.753702920016927,
|
| 1044 |
+
"eval_accuracy": 0.8842674042668678,
|
| 1045 |
+
"eval_loss": 0.5329333543777466,
|
| 1046 |
+
"eval_runtime": 56.7544,
|
| 1047 |
+
"eval_samples_per_second": 332.961,
|
| 1048 |
+
"eval_steps_per_second": 10.413,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 13.965298349555649,
|
| 1053 |
+
"grad_norm": 1.7780423164367676,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.5382,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 13.965298349555649,
|
| 1060 |
+
"eval_accuracy": 0.8849782989799418,
|
| 1061 |
+
"eval_loss": 0.5288933515548706,
|
| 1062 |
+
"eval_runtime": 56.7497,
|
| 1063 |
+
"eval_samples_per_second": 332.989,
|
| 1064 |
+
"eval_steps_per_second": 10.414,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 14.176893779094371,
|
| 1069 |
+
"grad_norm": 1.8220763206481934,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.5246,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 14.176893779094371,
|
| 1076 |
+
"eval_accuracy": 0.8849070057358006,
|
| 1077 |
+
"eval_loss": 0.5291073322296143,
|
| 1078 |
+
"eval_runtime": 56.71,
|
| 1079 |
+
"eval_samples_per_second": 333.221,
|
| 1080 |
+
"eval_steps_per_second": 10.421,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 14.388489208633093,
|
| 1085 |
+
"grad_norm": 2.090221881866455,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.5286,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 14.388489208633093,
|
| 1092 |
+
"eval_accuracy": 0.8858641979191566,
|
| 1093 |
+
"eval_loss": 0.5264419913291931,
|
| 1094 |
+
"eval_runtime": 56.7389,
|
| 1095 |
+
"eval_samples_per_second": 333.052,
|
| 1096 |
+
"eval_steps_per_second": 10.416,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 14.600084638171815,
|
| 1101 |
+
"grad_norm": 1.922879934310913,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.5276,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 14.600084638171815,
|
| 1108 |
+
"eval_accuracy": 0.8861009981828277,
|
| 1109 |
+
"eval_loss": 0.5238984227180481,
|
| 1110 |
+
"eval_runtime": 56.7406,
|
| 1111 |
+
"eval_samples_per_second": 333.042,
|
| 1112 |
+
"eval_steps_per_second": 10.416,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 14.811680067710537,
|
| 1117 |
+
"grad_norm": 1.8558276891708374,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.5257,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 14.811680067710537,
|
| 1124 |
+
"eval_accuracy": 0.887035860204965,
|
| 1125 |
+
"eval_loss": 0.5183533430099487,
|
| 1126 |
+
"eval_runtime": 56.8885,
|
| 1127 |
+
"eval_samples_per_second": 332.176,
|
| 1128 |
+
"eval_steps_per_second": 10.389,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 15.023275497249259,
|
| 1133 |
+
"grad_norm": 1.7486469745635986,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.5222,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 15.023275497249259,
|
| 1140 |
+
"eval_accuracy": 0.8865300123397599,
|
| 1141 |
+
"eval_loss": 0.5189926028251648,
|
| 1142 |
+
"eval_runtime": 56.7647,
|
| 1143 |
+
"eval_samples_per_second": 332.9,
|
| 1144 |
+
"eval_steps_per_second": 10.411,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 15.23487092678798,
|
| 1149 |
+
"grad_norm": 1.915968894958496,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.5154,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 15.23487092678798,
|
| 1156 |
+
"eval_accuracy": 0.8869442531931956,
|
| 1157 |
+
"eval_loss": 0.5159398913383484,
|
| 1158 |
+
"eval_runtime": 56.758,
|
| 1159 |
+
"eval_samples_per_second": 332.94,
|
| 1160 |
+
"eval_steps_per_second": 10.413,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 15.446466356326704,
|
| 1165 |
+
"grad_norm": 1.8159475326538086,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.5167,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 15.446466356326704,
|
| 1172 |
+
"eval_accuracy": 0.8873171814419585,
|
| 1173 |
+
"eval_loss": 0.5236589312553406,
|
| 1174 |
+
"eval_runtime": 56.7868,
|
| 1175 |
+
"eval_samples_per_second": 332.771,
|
| 1176 |
+
"eval_steps_per_second": 10.407,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 15.658061785865426,
|
| 1181 |
+
"grad_norm": 1.8895494937896729,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.5162,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 15.658061785865426,
|
| 1188 |
+
"eval_accuracy": 0.8873685835536078,
|
| 1189 |
+
"eval_loss": 0.5180982947349548,
|
| 1190 |
+
"eval_runtime": 56.8307,
|
| 1191 |
+
"eval_samples_per_second": 332.514,
|
| 1192 |
+
"eval_steps_per_second": 10.399,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 15.869657215404148,
|
| 1197 |
+
"grad_norm": 1.7780749797821045,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.5123,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 15.869657215404148,
|
| 1204 |
+
"eval_accuracy": 0.888074019743635,
|
| 1205 |
+
"eval_loss": 0.513282299041748,
|
| 1206 |
+
"eval_runtime": 56.7636,
|
| 1207 |
+
"eval_samples_per_second": 332.907,
|
| 1208 |
+
"eval_steps_per_second": 10.412,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 16.081252644942868,
|
| 1213 |
+
"grad_norm": 1.983821988105774,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.5152,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 16.081252644942868,
|
| 1220 |
+
"eval_accuracy": 0.8880291621546753,
|
| 1221 |
+
"eval_loss": 0.5154702663421631,
|
| 1222 |
+
"eval_runtime": 56.7842,
|
| 1223 |
+
"eval_samples_per_second": 332.786,
|
| 1224 |
+
"eval_steps_per_second": 10.408,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 16.29284807448159,
|
| 1229 |
+
"grad_norm": 1.695178747177124,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.5081,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 16.29284807448159,
|
| 1236 |
+
"eval_accuracy": 0.888959056609865,
|
| 1237 |
+
"eval_loss": 0.5113677382469177,
|
| 1238 |
+
"eval_runtime": 56.7684,
|
| 1239 |
+
"eval_samples_per_second": 332.879,
|
| 1240 |
+
"eval_steps_per_second": 10.411,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 16.504443504020312,
|
| 1245 |
+
"grad_norm": 1.6058467626571655,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.5078,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 16.504443504020312,
|
| 1252 |
+
"eval_accuracy": 0.8894830336569707,
|
| 1253 |
+
"eval_loss": 0.5070997476577759,
|
| 1254 |
+
"eval_runtime": 56.7413,
|
| 1255 |
+
"eval_samples_per_second": 333.038,
|
| 1256 |
+
"eval_steps_per_second": 10.416,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 16.716038933559034,
|
| 1261 |
+
"grad_norm": 1.764651894569397,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.5057,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 16.716038933559034,
|
| 1268 |
+
"eval_accuracy": 0.8894720089251245,
|
| 1269 |
+
"eval_loss": 0.5076988339424133,
|
| 1270 |
+
"eval_runtime": 56.7503,
|
| 1271 |
+
"eval_samples_per_second": 332.985,
|
| 1272 |
+
"eval_steps_per_second": 10.414,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 16.927634363097756,
|
| 1277 |
+
"grad_norm": 1.7007386684417725,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.512,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 16.927634363097756,
|
| 1284 |
+
"eval_accuracy": 0.8894052307731025,
|
| 1285 |
+
"eval_loss": 0.5073001384735107,
|
| 1286 |
+
"eval_runtime": 56.7848,
|
| 1287 |
+
"eval_samples_per_second": 332.783,
|
| 1288 |
+
"eval_steps_per_second": 10.408,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 17.139229792636478,
|
| 1293 |
+
"grad_norm": 1.8018271923065186,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.4984,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 17.139229792636478,
|
| 1300 |
+
"eval_accuracy": 0.8894733688436443,
|
| 1301 |
+
"eval_loss": 0.5062060952186584,
|
| 1302 |
+
"eval_runtime": 56.761,
|
| 1303 |
+
"eval_samples_per_second": 332.922,
|
| 1304 |
+
"eval_steps_per_second": 10.412,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 17.3508252221752,
|
| 1309 |
+
"grad_norm": 3.568917989730835,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.4954,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 17.3508252221752,
|
| 1316 |
+
"eval_accuracy": 0.8895164200933968,
|
| 1317 |
+
"eval_loss": 0.5070956349372864,
|
| 1318 |
+
"eval_runtime": 56.7457,
|
| 1319 |
+
"eval_samples_per_second": 333.012,
|
| 1320 |
+
"eval_steps_per_second": 10.415,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 17.56242065171392,
|
| 1325 |
+
"grad_norm": 1.7108345031738281,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.5003,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 17.56242065171392,
|
| 1332 |
+
"eval_accuracy": 0.8900638264840111,
|
| 1333 |
+
"eval_loss": 0.5031208395957947,
|
| 1334 |
+
"eval_runtime": 56.9117,
|
| 1335 |
+
"eval_samples_per_second": 332.041,
|
| 1336 |
+
"eval_steps_per_second": 10.385,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 17.774016081252643,
|
| 1341 |
+
"grad_norm": 1.6684004068374634,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.4979,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 17.774016081252643,
|
| 1348 |
+
"eval_accuracy": 0.8903730588680121,
|
| 1349 |
+
"eval_loss": 0.5011306405067444,
|
| 1350 |
+
"eval_runtime": 56.7247,
|
| 1351 |
+
"eval_samples_per_second": 333.135,
|
| 1352 |
+
"eval_steps_per_second": 10.419,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 17.985611510791365,
|
| 1357 |
+
"grad_norm": 2.0416200160980225,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.4979,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 17.985611510791365,
|
| 1364 |
+
"eval_accuracy": 0.8908411743071768,
|
| 1365 |
+
"eval_loss": 0.5000495910644531,
|
| 1366 |
+
"eval_runtime": 56.736,
|
| 1367 |
+
"eval_samples_per_second": 333.069,
|
| 1368 |
+
"eval_steps_per_second": 10.417,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 18.197206940330087,
|
| 1373 |
+
"grad_norm": 1.7811120748519897,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.49,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 18.197206940330087,
|
| 1380 |
+
"eval_accuracy": 0.8909345525264201,
|
| 1381 |
+
"eval_loss": 0.4998834729194641,
|
| 1382 |
+
"eval_runtime": 56.8954,
|
| 1383 |
+
"eval_samples_per_second": 332.136,
|
| 1384 |
+
"eval_steps_per_second": 10.387,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 18.408802369868813,
|
| 1389 |
+
"grad_norm": 1.8053951263427734,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.4885,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 18.408802369868813,
|
| 1396 |
+
"eval_accuracy": 0.8916438090401059,
|
| 1397 |
+
"eval_loss": 0.49805203080177307,
|
| 1398 |
+
"eval_runtime": 56.7641,
|
| 1399 |
+
"eval_samples_per_second": 332.904,
|
| 1400 |
+
"eval_steps_per_second": 10.412,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 18.620397799407534,
|
| 1405 |
+
"grad_norm": 1.7072595357894897,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.4927,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 18.620397799407534,
|
| 1412 |
+
"eval_accuracy": 0.8907857895790036,
|
| 1413 |
+
"eval_loss": 0.5022043585777283,
|
| 1414 |
+
"eval_runtime": 56.7625,
|
| 1415 |
+
"eval_samples_per_second": 332.914,
|
| 1416 |
+
"eval_steps_per_second": 10.412,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 18.831993228946256,
|
| 1421 |
+
"grad_norm": 1.6289252042770386,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.4837,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 18.831993228946256,
|
| 1428 |
+
"eval_accuracy": 0.8920152847579453,
|
| 1429 |
+
"eval_loss": 0.49153533577919006,
|
| 1430 |
+
"eval_runtime": 56.7452,
|
| 1431 |
+
"eval_samples_per_second": 333.015,
|
| 1432 |
+
"eval_steps_per_second": 10.415,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 19.04358865848498,
|
| 1437 |
+
"grad_norm": 1.8188107013702393,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.4837,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 19.04358865848498,
|
| 1444 |
+
"eval_accuracy": 0.8912697247648497,
|
| 1445 |
+
"eval_loss": 0.5010623335838318,
|
| 1446 |
+
"eval_runtime": 56.7313,
|
| 1447 |
+
"eval_samples_per_second": 333.097,
|
| 1448 |
+
"eval_steps_per_second": 10.418,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 19.2551840880237,
|
| 1453 |
+
"grad_norm": 1.9126912355422974,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.4859,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 19.2551840880237,
|
| 1460 |
+
"eval_accuracy": 0.8923426689616218,
|
| 1461 |
+
"eval_loss": 0.49420884251594543,
|
| 1462 |
+
"eval_runtime": 56.7991,
|
| 1463 |
+
"eval_samples_per_second": 332.699,
|
| 1464 |
+
"eval_steps_per_second": 10.405,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 19.466779517562422,
|
| 1469 |
+
"grad_norm": 2.0371763706207275,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.4805,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 19.466779517562422,
|
| 1476 |
+
"eval_accuracy": 0.8919669118661465,
|
| 1477 |
+
"eval_loss": 0.49629610776901245,
|
| 1478 |
+
"eval_runtime": 56.7198,
|
| 1479 |
+
"eval_samples_per_second": 333.164,
|
| 1480 |
+
"eval_steps_per_second": 10.42,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 19.678374947101144,
|
| 1485 |
+
"grad_norm": 1.6247458457946777,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.482,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 19.678374947101144,
|
| 1492 |
+
"eval_accuracy": 0.8925763545595373,
|
| 1493 |
+
"eval_loss": 0.4910178780555725,
|
| 1494 |
+
"eval_runtime": 56.7594,
|
| 1495 |
+
"eval_samples_per_second": 332.931,
|
| 1496 |
+
"eval_steps_per_second": 10.412,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 19.889970376639866,
|
| 1501 |
+
"grad_norm": 1.7761908769607544,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.4795,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 19.889970376639866,
|
| 1508 |
+
"eval_accuracy": 0.8927224378040853,
|
| 1509 |
+
"eval_loss": 0.49156272411346436,
|
| 1510 |
+
"eval_runtime": 56.7677,
|
| 1511 |
+
"eval_samples_per_second": 332.883,
|
| 1512 |
+
"eval_steps_per_second": 10.411,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 20.101565806178588,
|
| 1517 |
+
"grad_norm": 1.7788410186767578,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.4768,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 20.101565806178588,
|
| 1524 |
+
"eval_accuracy": 0.8934505144730001,
|
| 1525 |
+
"eval_loss": 0.49061286449432373,
|
| 1526 |
+
"eval_runtime": 56.7712,
|
| 1527 |
+
"eval_samples_per_second": 332.863,
|
| 1528 |
+
"eval_steps_per_second": 10.41,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 20.31316123571731,
|
| 1533 |
+
"grad_norm": 1.5896297693252563,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.4742,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 20.31316123571731,
|
| 1540 |
+
"eval_accuracy": 0.8928689756013872,
|
| 1541 |
+
"eval_loss": 0.48972323536872864,
|
| 1542 |
+
"eval_runtime": 56.8531,
|
| 1543 |
+
"eval_samples_per_second": 332.383,
|
| 1544 |
+
"eval_steps_per_second": 10.395,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 20.52475666525603,
|
| 1549 |
+
"grad_norm": 1.7268341779708862,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.4745,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 20.52475666525603,
|
| 1556 |
+
"eval_accuracy": 0.8931779879053388,
|
| 1557 |
+
"eval_loss": 0.48947346210479736,
|
| 1558 |
+
"eval_runtime": 56.7482,
|
| 1559 |
+
"eval_samples_per_second": 332.997,
|
| 1560 |
+
"eval_steps_per_second": 10.414,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 20.736352094794753,
|
| 1565 |
+
"grad_norm": 1.5962858200073242,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.4762,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 20.736352094794753,
|
| 1572 |
+
"eval_accuracy": 0.8939185209075733,
|
| 1573 |
+
"eval_loss": 0.48672759532928467,
|
| 1574 |
+
"eval_runtime": 56.7164,
|
| 1575 |
+
"eval_samples_per_second": 333.184,
|
| 1576 |
+
"eval_steps_per_second": 10.42,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 20.947947524333475,
|
| 1581 |
+
"grad_norm": 1.7846648693084717,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.4762,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 20.947947524333475,
|
| 1588 |
+
"eval_accuracy": 0.8931934715783109,
|
| 1589 |
+
"eval_loss": 0.48964911699295044,
|
| 1590 |
+
"eval_runtime": 56.7779,
|
| 1591 |
+
"eval_samples_per_second": 332.823,
|
| 1592 |
+
"eval_steps_per_second": 10.409,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 21.159542953872197,
|
| 1597 |
+
"grad_norm": 1.8739550113677979,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.4697,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 21.159542953872197,
|
| 1604 |
+
"eval_accuracy": 0.8942228433369652,
|
| 1605 |
+
"eval_loss": 0.4863516390323639,
|
| 1606 |
+
"eval_runtime": 56.7879,
|
| 1607 |
+
"eval_samples_per_second": 332.765,
|
| 1608 |
+
"eval_steps_per_second": 10.407,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 21.37113838341092,
|
| 1613 |
+
"grad_norm": 1.6317665576934814,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.4676,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 21.37113838341092,
|
| 1620 |
+
"eval_accuracy": 0.8943121055442946,
|
| 1621 |
+
"eval_loss": 0.48264408111572266,
|
| 1622 |
+
"eval_runtime": 56.7499,
|
| 1623 |
+
"eval_samples_per_second": 332.987,
|
| 1624 |
+
"eval_steps_per_second": 10.414,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 21.58273381294964,
|
| 1629 |
+
"grad_norm": 1.8186049461364746,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.4691,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 21.58273381294964,
|
| 1636 |
+
"eval_accuracy": 0.8941176840592336,
|
| 1637 |
+
"eval_loss": 0.48747971653938293,
|
| 1638 |
+
"eval_runtime": 56.7807,
|
| 1639 |
+
"eval_samples_per_second": 332.806,
|
| 1640 |
+
"eval_steps_per_second": 10.408,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 21.794329242488363,
|
| 1645 |
+
"grad_norm": 1.67414128780365,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.4662,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 21.794329242488363,
|
| 1652 |
+
"eval_accuracy": 0.8945745628903783,
|
| 1653 |
+
"eval_loss": 0.48337647318840027,
|
| 1654 |
+
"eval_runtime": 55.7176,
|
| 1655 |
+
"eval_samples_per_second": 339.156,
|
| 1656 |
+
"eval_steps_per_second": 10.607,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 22.005924672027085,
|
| 1661 |
+
"grad_norm": 1.4717955589294434,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.4634,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 22.005924672027085,
|
| 1668 |
+
"eval_accuracy": 0.8949934626709839,
|
| 1669 |
+
"eval_loss": 0.4821498394012451,
|
| 1670 |
+
"eval_runtime": 56.7634,
|
| 1671 |
+
"eval_samples_per_second": 332.908,
|
| 1672 |
+
"eval_steps_per_second": 10.412,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 22.217520101565807,
|
| 1677 |
+
"grad_norm": 1.7725749015808105,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.4547,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 22.217520101565807,
|
| 1684 |
+
"eval_accuracy": 0.8948994692290997,
|
| 1685 |
+
"eval_loss": 0.4810987114906311,
|
| 1686 |
+
"eval_runtime": 56.7441,
|
| 1687 |
+
"eval_samples_per_second": 333.021,
|
| 1688 |
+
"eval_steps_per_second": 10.415,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 22.42911553110453,
|
| 1693 |
+
"grad_norm": 1.669144630432129,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.4558,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 22.42911553110453,
|
| 1700 |
+
"eval_accuracy": 0.8952562835440645,
|
| 1701 |
+
"eval_loss": 0.4825162887573242,
|
| 1702 |
+
"eval_runtime": 56.8582,
|
| 1703 |
+
"eval_samples_per_second": 332.353,
|
| 1704 |
+
"eval_steps_per_second": 10.394,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 22.64071096064325,
|
| 1709 |
+
"grad_norm": 1.6026239395141602,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.4621,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 22.64071096064325,
|
| 1716 |
+
"eval_accuracy": 0.8955995271171235,
|
| 1717 |
+
"eval_loss": 0.47604426741600037,
|
| 1718 |
+
"eval_runtime": 56.7769,
|
| 1719 |
+
"eval_samples_per_second": 332.829,
|
| 1720 |
+
"eval_steps_per_second": 10.409,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 22.852306390181973,
|
| 1725 |
+
"grad_norm": 2.0116236209869385,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.4617,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 22.852306390181973,
|
| 1732 |
+
"eval_accuracy": 0.8959320153668214,
|
| 1733 |
+
"eval_loss": 0.4762740433216095,
|
| 1734 |
+
"eval_runtime": 56.7895,
|
| 1735 |
+
"eval_samples_per_second": 332.755,
|
| 1736 |
+
"eval_steps_per_second": 10.407,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 23.063901819720694,
|
| 1741 |
+
"grad_norm": 1.9537534713745117,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.4547,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 23.063901819720694,
|
| 1748 |
+
"eval_accuracy": 0.8956957929057274,
|
| 1749 |
+
"eval_loss": 0.4776069223880768,
|
| 1750 |
+
"eval_runtime": 56.7746,
|
| 1751 |
+
"eval_samples_per_second": 332.842,
|
| 1752 |
+
"eval_steps_per_second": 10.41,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 23.275497249259416,
|
| 1757 |
+
"grad_norm": 1.838956594467163,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.4499,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 23.275497249259416,
|
| 1764 |
+
"eval_accuracy": 0.8959442254178052,
|
| 1765 |
+
"eval_loss": 0.4744105637073517,
|
| 1766 |
+
"eval_runtime": 56.7338,
|
| 1767 |
+
"eval_samples_per_second": 333.082,
|
| 1768 |
+
"eval_steps_per_second": 10.417,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 23.487092678798138,
|
| 1773 |
+
"grad_norm": 1.6302268505096436,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.4561,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 23.487092678798138,
|
| 1780 |
+
"eval_accuracy": 0.8958769277243019,
|
| 1781 |
+
"eval_loss": 0.4771183431148529,
|
| 1782 |
+
"eval_runtime": 56.7479,
|
| 1783 |
+
"eval_samples_per_second": 332.999,
|
| 1784 |
+
"eval_steps_per_second": 10.414,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 23.69868810833686,
|
| 1789 |
+
"grad_norm": 1.8881497383117676,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.452,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 23.69868810833686,
|
| 1796 |
+
"eval_accuracy": 0.8964990723620592,
|
| 1797 |
+
"eval_loss": 0.4719561040401459,
|
| 1798 |
+
"eval_runtime": 56.7228,
|
| 1799 |
+
"eval_samples_per_second": 333.147,
|
| 1800 |
+
"eval_steps_per_second": 10.419,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 23.910283537875582,
|
| 1805 |
+
"grad_norm": 1.4877843856811523,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.4501,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 23.910283537875582,
|
| 1812 |
+
"eval_accuracy": 0.897199972583508,
|
| 1813 |
+
"eval_loss": 0.471039742231369,
|
| 1814 |
+
"eval_runtime": 56.7779,
|
| 1815 |
+
"eval_samples_per_second": 332.823,
|
| 1816 |
+
"eval_steps_per_second": 10.409,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 24.121878967414304,
|
| 1821 |
+
"grad_norm": 1.7829620838165283,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.4455,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 24.121878967414304,
|
| 1828 |
+
"eval_accuracy": 0.8968278913027552,
|
| 1829 |
+
"eval_loss": 0.4751090109348297,
|
| 1830 |
+
"eval_runtime": 56.7642,
|
| 1831 |
+
"eval_samples_per_second": 332.904,
|
| 1832 |
+
"eval_steps_per_second": 10.411,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 24.333474396953026,
|
| 1837 |
+
"grad_norm": 1.576822280883789,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.4435,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 24.333474396953026,
|
| 1844 |
+
"eval_accuracy": 0.8976774423621006,
|
| 1845 |
+
"eval_loss": 0.46997109055519104,
|
| 1846 |
+
"eval_runtime": 56.7564,
|
| 1847 |
+
"eval_samples_per_second": 332.949,
|
| 1848 |
+
"eval_steps_per_second": 10.413,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 24.545069826491748,
|
| 1853 |
+
"grad_norm": 1.5602706670761108,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.4474,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 24.545069826491748,
|
| 1860 |
+
"eval_accuracy": 0.8972810308423196,
|
| 1861 |
+
"eval_loss": 0.4720916152000427,
|
| 1862 |
+
"eval_runtime": 56.7869,
|
| 1863 |
+
"eval_samples_per_second": 332.77,
|
| 1864 |
+
"eval_steps_per_second": 10.407,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 24.75666525603047,
|
| 1869 |
+
"grad_norm": 1.9054083824157715,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.4444,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 24.75666525603047,
|
| 1876 |
+
"eval_accuracy": 0.8969371251644085,
|
| 1877 |
+
"eval_loss": 0.47275567054748535,
|
| 1878 |
+
"eval_runtime": 56.7839,
|
| 1879 |
+
"eval_samples_per_second": 332.788,
|
| 1880 |
+
"eval_steps_per_second": 10.408,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 24.96826068556919,
|
| 1885 |
+
"grad_norm": 1.5889158248901367,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.4505,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 24.96826068556919,
|
| 1892 |
+
"eval_accuracy": 0.8973650392195069,
|
| 1893 |
+
"eval_loss": 0.4704943597316742,
|
| 1894 |
+
"eval_runtime": 56.8826,
|
| 1895 |
+
"eval_samples_per_second": 332.211,
|
| 1896 |
+
"eval_steps_per_second": 10.39,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 25.179856115107913,
|
| 1901 |
+
"grad_norm": 1.7696033716201782,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.4385,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 25.179856115107913,
|
| 1908 |
+
"eval_accuracy": 0.8985368520733181,
|
| 1909 |
+
"eval_loss": 0.4653254449367523,
|
| 1910 |
+
"eval_runtime": 56.7868,
|
| 1911 |
+
"eval_samples_per_second": 332.771,
|
| 1912 |
+
"eval_steps_per_second": 10.407,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 25.391451544646635,
|
| 1917 |
+
"grad_norm": 1.7337836027145386,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.4402,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 25.391451544646635,
|
| 1924 |
+
"eval_accuracy": 0.8977789289543835,
|
| 1925 |
+
"eval_loss": 0.4724612236022949,
|
| 1926 |
+
"eval_runtime": 56.7658,
|
| 1927 |
+
"eval_samples_per_second": 332.894,
|
| 1928 |
+
"eval_steps_per_second": 10.411,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 25.603046974185357,
|
| 1933 |
+
"grad_norm": 1.5213319063186646,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.4377,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 25.603046974185357,
|
| 1940 |
+
"eval_accuracy": 0.8985293051778416,
|
| 1941 |
+
"eval_loss": 0.4643355906009674,
|
| 1942 |
+
"eval_runtime": 56.7771,
|
| 1943 |
+
"eval_samples_per_second": 332.828,
|
| 1944 |
+
"eval_steps_per_second": 10.409,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 25.81464240372408,
|
| 1949 |
+
"grad_norm": 1.6032094955444336,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.4408,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 25.81464240372408,
|
| 1956 |
+
"eval_accuracy": 0.8980092011578702,
|
| 1957 |
+
"eval_loss": 0.4661051034927368,
|
| 1958 |
+
"eval_runtime": 56.7636,
|
| 1959 |
+
"eval_samples_per_second": 332.907,
|
| 1960 |
+
"eval_steps_per_second": 10.412,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 26.0262378332628,
|
| 1965 |
+
"grad_norm": 1.5952712297439575,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.4402,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 26.0262378332628,
|
| 1972 |
+
"eval_accuracy": 0.8991291446824378,
|
| 1973 |
+
"eval_loss": 0.45936518907546997,
|
| 1974 |
+
"eval_runtime": 56.7912,
|
| 1975 |
+
"eval_samples_per_second": 332.745,
|
| 1976 |
+
"eval_steps_per_second": 10.407,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 26.237833262801523,
|
| 1981 |
+
"grad_norm": 1.7169263362884521,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.4359,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 26.237833262801523,
|
| 1988 |
+
"eval_accuracy": 0.8993624862013838,
|
| 1989 |
+
"eval_loss": 0.4599079191684723,
|
| 1990 |
+
"eval_runtime": 56.7617,
|
| 1991 |
+
"eval_samples_per_second": 332.918,
|
| 1992 |
+
"eval_steps_per_second": 10.412,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 26.449428692340245,
|
| 1997 |
+
"grad_norm": 1.5549594163894653,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.434,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 26.449428692340245,
|
| 2004 |
+
"eval_accuracy": 0.898992199555206,
|
| 2005 |
+
"eval_loss": 0.4616684913635254,
|
| 2006 |
+
"eval_runtime": 56.7929,
|
| 2007 |
+
"eval_samples_per_second": 332.735,
|
| 2008 |
+
"eval_steps_per_second": 10.406,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 26.661024121878967,
|
| 2013 |
+
"grad_norm": 1.5618722438812256,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.4322,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 26.661024121878967,
|
| 2020 |
+
"eval_accuracy": 0.8992191360919601,
|
| 2021 |
+
"eval_loss": 0.4595487415790558,
|
| 2022 |
+
"eval_runtime": 56.7516,
|
| 2023 |
+
"eval_samples_per_second": 332.978,
|
| 2024 |
+
"eval_steps_per_second": 10.414,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 26.87261955141769,
|
| 2029 |
+
"grad_norm": 1.4580327272415161,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.4326,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 26.87261955141769,
|
| 2036 |
+
"eval_accuracy": 0.8998144186324953,
|
| 2037 |
+
"eval_loss": 0.45802512764930725,
|
| 2038 |
+
"eval_runtime": 56.7668,
|
| 2039 |
+
"eval_samples_per_second": 332.888,
|
| 2040 |
+
"eval_steps_per_second": 10.411,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 27.08421498095641,
|
| 2045 |
+
"grad_norm": 1.5803847312927246,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.4292,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 27.08421498095641,
|
| 2052 |
+
"eval_accuracy": 0.9003762227238525,
|
| 2053 |
+
"eval_loss": 0.45433345437049866,
|
| 2054 |
+
"eval_runtime": 56.8145,
|
| 2055 |
+
"eval_samples_per_second": 332.609,
|
| 2056 |
+
"eval_steps_per_second": 10.402,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 27.295810410495132,
|
| 2061 |
+
"grad_norm": 1.7439405918121338,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.4273,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 27.295810410495132,
|
| 2068 |
+
"eval_accuracy": 0.8997829988720355,
|
| 2069 |
+
"eval_loss": 0.4542884826660156,
|
| 2070 |
+
"eval_runtime": 56.757,
|
| 2071 |
+
"eval_samples_per_second": 332.946,
|
| 2072 |
+
"eval_steps_per_second": 10.413,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 27.507405840033854,
|
| 2077 |
+
"grad_norm": 1.5799704790115356,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.4271,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 27.507405840033854,
|
| 2084 |
+
"eval_accuracy": 0.9005278512106127,
|
| 2085 |
+
"eval_loss": 0.4549398720264435,
|
| 2086 |
+
"eval_runtime": 56.7761,
|
| 2087 |
+
"eval_samples_per_second": 332.834,
|
| 2088 |
+
"eval_steps_per_second": 10.409,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 27.719001269572576,
|
| 2093 |
+
"grad_norm": 1.671703815460205,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.4225,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 27.719001269572576,
|
| 2100 |
+
"eval_accuracy": 0.9003234757377444,
|
| 2101 |
+
"eval_loss": 0.4551466405391693,
|
| 2102 |
+
"eval_runtime": 56.7802,
|
| 2103 |
+
"eval_samples_per_second": 332.81,
|
| 2104 |
+
"eval_steps_per_second": 10.409,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 27.930596699111298,
|
| 2109 |
+
"grad_norm": 1.6339532136917114,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.4265,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 27.930596699111298,
|
| 2116 |
+
"eval_accuracy": 0.9002172699273693,
|
| 2117 |
+
"eval_loss": 0.4552587866783142,
|
| 2118 |
+
"eval_runtime": 56.7694,
|
| 2119 |
+
"eval_samples_per_second": 332.873,
|
| 2120 |
+
"eval_steps_per_second": 10.411,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 28.14219212865002,
|
| 2125 |
+
"grad_norm": 1.8030368089675903,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.4244,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 28.14219212865002,
|
| 2132 |
+
"eval_accuracy": 0.9006673602396534,
|
| 2133 |
+
"eval_loss": 0.4554223418235779,
|
| 2134 |
+
"eval_runtime": 56.7481,
|
| 2135 |
+
"eval_samples_per_second": 332.998,
|
| 2136 |
+
"eval_steps_per_second": 10.414,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 28.353787558188742,
|
| 2141 |
+
"grad_norm": 1.6583260297775269,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.422,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 28.353787558188742,
|
| 2148 |
+
"eval_accuracy": 0.9013144719641863,
|
| 2149 |
+
"eval_loss": 0.45334717631340027,
|
| 2150 |
+
"eval_runtime": 56.7592,
|
| 2151 |
+
"eval_samples_per_second": 332.933,
|
| 2152 |
+
"eval_steps_per_second": 10.412,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 28.565382987727464,
|
| 2157 |
+
"grad_norm": 1.6311790943145752,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.4188,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 28.565382987727464,
|
| 2164 |
+
"eval_accuracy": 0.9004746460616811,
|
| 2165 |
+
"eval_loss": 0.4538644254207611,
|
| 2166 |
+
"eval_runtime": 56.7734,
|
| 2167 |
+
"eval_samples_per_second": 332.85,
|
| 2168 |
+
"eval_steps_per_second": 10.41,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 28.776978417266186,
|
| 2173 |
+
"grad_norm": 1.8188596963882446,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.4236,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 28.776978417266186,
|
| 2180 |
+
"eval_accuracy": 0.9012951971056801,
|
| 2181 |
+
"eval_loss": 0.4524097144603729,
|
| 2182 |
+
"eval_runtime": 56.7821,
|
| 2183 |
+
"eval_samples_per_second": 332.799,
|
| 2184 |
+
"eval_steps_per_second": 10.408,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 28.988573846804908,
|
| 2189 |
+
"grad_norm": 1.7251187562942505,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.4202,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 28.988573846804908,
|
| 2196 |
+
"eval_accuracy": 0.9014573223974446,
|
| 2197 |
+
"eval_loss": 0.44895613193511963,
|
| 2198 |
+
"eval_runtime": 56.8737,
|
| 2199 |
+
"eval_samples_per_second": 332.263,
|
| 2200 |
+
"eval_steps_per_second": 10.391,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 29.20016927634363,
|
| 2205 |
+
"grad_norm": 1.7557555437088013,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.419,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 29.20016927634363,
|
| 2212 |
+
"eval_accuracy": 0.9017764579643729,
|
| 2213 |
+
"eval_loss": 0.4491879940032959,
|
| 2214 |
+
"eval_runtime": 56.7246,
|
| 2215 |
+
"eval_samples_per_second": 333.136,
|
| 2216 |
+
"eval_steps_per_second": 10.419,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 29.41176470588235,
|
| 2221 |
+
"grad_norm": 1.68495774269104,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.4159,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 29.41176470588235,
|
| 2228 |
+
"eval_accuracy": 0.9013985887304796,
|
| 2229 |
+
"eval_loss": 0.45250919461250305,
|
| 2230 |
+
"eval_runtime": 56.8205,
|
| 2231 |
+
"eval_samples_per_second": 332.574,
|
| 2232 |
+
"eval_steps_per_second": 10.401,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 29.623360135421073,
|
| 2237 |
+
"grad_norm": 1.6274399757385254,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.416,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 29.623360135421073,
|
| 2244 |
+
"eval_accuracy": 0.9011979465572715,
|
| 2245 |
+
"eval_loss": 0.45176440477371216,
|
| 2246 |
+
"eval_runtime": 56.7549,
|
| 2247 |
+
"eval_samples_per_second": 332.958,
|
| 2248 |
+
"eval_steps_per_second": 10.413,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 29.834955564959795,
|
| 2253 |
+
"grad_norm": 1.6035505533218384,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.4152,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 29.834955564959795,
|
| 2260 |
+
"eval_accuracy": 0.9019880801849817,
|
| 2261 |
+
"eval_loss": 0.4507512152194977,
|
| 2262 |
+
"eval_runtime": 56.8264,
|
| 2263 |
+
"eval_samples_per_second": 332.539,
|
| 2264 |
+
"eval_steps_per_second": 10.4,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 30.046550994498517,
|
| 2269 |
+
"grad_norm": 1.8008580207824707,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.4172,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 30.046550994498517,
|
| 2276 |
+
"eval_accuracy": 0.9021050974217674,
|
| 2277 |
+
"eval_loss": 0.4502410888671875,
|
| 2278 |
+
"eval_runtime": 56.7561,
|
| 2279 |
+
"eval_samples_per_second": 332.951,
|
| 2280 |
+
"eval_steps_per_second": 10.413,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 30.25814642403724,
|
| 2285 |
+
"grad_norm": 2.0598397254943848,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.4118,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 30.25814642403724,
|
| 2292 |
+
"eval_accuracy": 0.9019547233430508,
|
| 2293 |
+
"eval_loss": 0.4497910439968109,
|
| 2294 |
+
"eval_runtime": 56.7678,
|
| 2295 |
+
"eval_samples_per_second": 332.882,
|
| 2296 |
+
"eval_steps_per_second": 10.411,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 30.46974185357596,
|
| 2301 |
+
"grad_norm": 1.6746747493743896,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.4081,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 30.46974185357596,
|
| 2308 |
+
"eval_accuracy": 0.9026417566458294,
|
| 2309 |
+
"eval_loss": 0.4457666575908661,
|
| 2310 |
+
"eval_runtime": 56.7772,
|
| 2311 |
+
"eval_samples_per_second": 332.827,
|
| 2312 |
+
"eval_steps_per_second": 10.409,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 30.681337283114686,
|
| 2317 |
+
"grad_norm": 1.6497710943222046,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.411,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 30.681337283114686,
|
| 2324 |
+
"eval_accuracy": 0.9021046343620955,
|
| 2325 |
+
"eval_loss": 0.4476597309112549,
|
| 2326 |
+
"eval_runtime": 56.7788,
|
| 2327 |
+
"eval_samples_per_second": 332.818,
|
| 2328 |
+
"eval_steps_per_second": 10.409,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 30.89293271265341,
|
| 2333 |
+
"grad_norm": 1.8102821111679077,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.409,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 30.89293271265341,
|
| 2340 |
+
"eval_accuracy": 0.9025454516766531,
|
| 2341 |
+
"eval_loss": 0.4461187422275543,
|
| 2342 |
+
"eval_runtime": 56.7848,
|
| 2343 |
+
"eval_samples_per_second": 332.783,
|
| 2344 |
+
"eval_steps_per_second": 10.408,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 31.10452814219213,
|
| 2349 |
+
"grad_norm": 1.58904230594635,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.4118,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 31.10452814219213,
|
| 2356 |
+
"eval_accuracy": 0.90234226951983,
|
| 2357 |
+
"eval_loss": 0.4481985569000244,
|
| 2358 |
+
"eval_runtime": 56.7757,
|
| 2359 |
+
"eval_samples_per_second": 332.836,
|
| 2360 |
+
"eval_steps_per_second": 10.409,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 31.316123571730852,
|
| 2365 |
+
"grad_norm": 1.7424213886260986,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.4022,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 31.316123571730852,
|
| 2372 |
+
"eval_accuracy": 0.9033274732451086,
|
| 2373 |
+
"eval_loss": 0.4427095651626587,
|
| 2374 |
+
"eval_runtime": 56.7818,
|
| 2375 |
+
"eval_samples_per_second": 332.8,
|
| 2376 |
+
"eval_steps_per_second": 10.408,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 31.527719001269574,
|
| 2381 |
+
"grad_norm": 1.766745686531067,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.4083,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 31.527719001269574,
|
| 2388 |
+
"eval_accuracy": 0.9030700219715095,
|
| 2389 |
+
"eval_loss": 0.4428587257862091,
|
| 2390 |
+
"eval_runtime": 56.7713,
|
| 2391 |
+
"eval_samples_per_second": 332.862,
|
| 2392 |
+
"eval_steps_per_second": 10.41,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 31.739314430808296,
|
| 2397 |
+
"grad_norm": 1.920547604560852,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.4084,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 31.739314430808296,
|
| 2404 |
+
"eval_accuracy": 0.9029642025071578,
|
| 2405 |
+
"eval_loss": 0.4436880350112915,
|
| 2406 |
+
"eval_runtime": 56.7524,
|
| 2407 |
+
"eval_samples_per_second": 332.973,
|
| 2408 |
+
"eval_steps_per_second": 10.414,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 31.950909860347018,
|
| 2413 |
+
"grad_norm": 1.5884405374526978,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.4029,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 31.950909860347018,
|
| 2420 |
+
"eval_accuracy": 0.9039806372927507,
|
| 2421 |
+
"eval_loss": 0.438906729221344,
|
| 2422 |
+
"eval_runtime": 56.7528,
|
| 2423 |
+
"eval_samples_per_second": 332.97,
|
| 2424 |
+
"eval_steps_per_second": 10.414,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 32.162505289885736,
|
| 2429 |
+
"grad_norm": 1.8102740049362183,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.4031,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 32.162505289885736,
|
| 2436 |
+
"eval_accuracy": 0.9035506519558676,
|
| 2437 |
+
"eval_loss": 0.4416000545024872,
|
| 2438 |
+
"eval_runtime": 56.7957,
|
| 2439 |
+
"eval_samples_per_second": 332.719,
|
| 2440 |
+
"eval_steps_per_second": 10.406,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 32.37410071942446,
|
| 2445 |
+
"grad_norm": 1.5819768905639648,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.403,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 32.37410071942446,
|
| 2452 |
+
"eval_accuracy": 0.9042462224232761,
|
| 2453 |
+
"eval_loss": 0.43676483631134033,
|
| 2454 |
+
"eval_runtime": 56.7531,
|
| 2455 |
+
"eval_samples_per_second": 332.969,
|
| 2456 |
+
"eval_steps_per_second": 10.414,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 32.58569614896318,
|
| 2461 |
+
"grad_norm": 1.566695213317871,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.4006,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 32.58569614896318,
|
| 2468 |
+
"eval_accuracy": 0.9042393613749178,
|
| 2469 |
+
"eval_loss": 0.4374808669090271,
|
| 2470 |
+
"eval_runtime": 56.7895,
|
| 2471 |
+
"eval_samples_per_second": 332.755,
|
| 2472 |
+
"eval_steps_per_second": 10.407,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 32.797291578501905,
|
| 2477 |
+
"grad_norm": 1.7661139965057373,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.402,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 32.797291578501905,
|
| 2484 |
+
"eval_accuracy": 0.903702221238192,
|
| 2485 |
+
"eval_loss": 0.4417952299118042,
|
| 2486 |
+
"eval_runtime": 56.7647,
|
| 2487 |
+
"eval_samples_per_second": 332.9,
|
| 2488 |
+
"eval_steps_per_second": 10.411,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 33.008887008040624,
|
| 2493 |
+
"grad_norm": 1.7405301332473755,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.4012,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 33.008887008040624,
|
| 2500 |
+
"eval_accuracy": 0.904089089850285,
|
| 2501 |
+
"eval_loss": 0.44198983907699585,
|
| 2502 |
+
"eval_runtime": 56.7736,
|
| 2503 |
+
"eval_samples_per_second": 332.848,
|
| 2504 |
+
"eval_steps_per_second": 10.41,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 33.22048243757935,
|
| 2509 |
+
"grad_norm": 1.682528018951416,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.3935,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 33.22048243757935,
|
| 2516 |
+
"eval_accuracy": 0.9044004264348643,
|
| 2517 |
+
"eval_loss": 0.43821093440055847,
|
| 2518 |
+
"eval_runtime": 56.78,
|
| 2519 |
+
"eval_samples_per_second": 332.811,
|
| 2520 |
+
"eval_steps_per_second": 10.409,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 33.43207786711807,
|
| 2525 |
+
"grad_norm": 1.7806745767593384,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.3935,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 33.43207786711807,
|
| 2532 |
+
"eval_accuracy": 0.904705222086893,
|
| 2533 |
+
"eval_loss": 0.43635934591293335,
|
| 2534 |
+
"eval_runtime": 56.7716,
|
| 2535 |
+
"eval_samples_per_second": 332.86,
|
| 2536 |
+
"eval_steps_per_second": 10.41,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 33.64367329665679,
|
| 2541 |
+
"grad_norm": 1.7414941787719727,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.3944,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 33.64367329665679,
|
| 2548 |
+
"eval_accuracy": 0.9044635372679648,
|
| 2549 |
+
"eval_loss": 0.436748743057251,
|
| 2550 |
+
"eval_runtime": 56.8872,
|
| 2551 |
+
"eval_samples_per_second": 332.184,
|
| 2552 |
+
"eval_steps_per_second": 10.389,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 33.85526872619551,
|
| 2557 |
+
"grad_norm": 1.5406990051269531,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.3974,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 33.85526872619551,
|
| 2564 |
+
"eval_accuracy": 0.9042456122364483,
|
| 2565 |
+
"eval_loss": 0.4392504394054413,
|
| 2566 |
+
"eval_runtime": 56.7872,
|
| 2567 |
+
"eval_samples_per_second": 332.769,
|
| 2568 |
+
"eval_steps_per_second": 10.407,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 34.06686415573424,
|
| 2573 |
+
"grad_norm": 1.705238938331604,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.3943,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 34.06686415573424,
|
| 2580 |
+
"eval_accuracy": 0.9046601613087342,
|
| 2581 |
+
"eval_loss": 0.43598660826683044,
|
| 2582 |
+
"eval_runtime": 56.8084,
|
| 2583 |
+
"eval_samples_per_second": 332.644,
|
| 2584 |
+
"eval_steps_per_second": 10.403,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 34.278459585272955,
|
| 2589 |
+
"grad_norm": 1.7163870334625244,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.3945,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 34.278459585272955,
|
| 2596 |
+
"eval_accuracy": 0.9047847765427016,
|
| 2597 |
+
"eval_loss": 0.4343646764755249,
|
| 2598 |
+
"eval_runtime": 56.8202,
|
| 2599 |
+
"eval_samples_per_second": 332.575,
|
| 2600 |
+
"eval_steps_per_second": 10.401,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 34.49005501481168,
|
| 2605 |
+
"grad_norm": 1.6511443853378296,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.3875,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 34.49005501481168,
|
| 2612 |
+
"eval_accuracy": 0.9054521754941371,
|
| 2613 |
+
"eval_loss": 0.43327948451042175,
|
| 2614 |
+
"eval_runtime": 56.8074,
|
| 2615 |
+
"eval_samples_per_second": 332.65,
|
| 2616 |
+
"eval_steps_per_second": 10.404,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 34.7016504443504,
|
| 2621 |
+
"grad_norm": 1.6507718563079834,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.3914,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 34.7016504443504,
|
| 2628 |
+
"eval_accuracy": 0.9054627570063298,
|
| 2629 |
+
"eval_loss": 0.43327319622039795,
|
| 2630 |
+
"eval_runtime": 56.7623,
|
| 2631 |
+
"eval_samples_per_second": 332.914,
|
| 2632 |
+
"eval_steps_per_second": 10.412,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 34.913245873889124,
|
| 2637 |
+
"grad_norm": 1.5144648551940918,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.3826,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 34.913245873889124,
|
| 2644 |
+
"eval_accuracy": 0.905242186429115,
|
| 2645 |
+
"eval_loss": 0.4331113398075104,
|
| 2646 |
+
"eval_runtime": 56.7962,
|
| 2647 |
+
"eval_samples_per_second": 332.716,
|
| 2648 |
+
"eval_steps_per_second": 10.406,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 35.12484130342784,
|
| 2653 |
+
"grad_norm": 1.760556697845459,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.3919,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 35.12484130342784,
|
| 2660 |
+
"eval_accuracy": 0.9060222399218201,
|
| 2661 |
+
"eval_loss": 0.429587721824646,
|
| 2662 |
+
"eval_runtime": 56.7833,
|
| 2663 |
+
"eval_samples_per_second": 332.792,
|
| 2664 |
+
"eval_steps_per_second": 10.408,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 35.33643673296657,
|
| 2669 |
+
"grad_norm": 1.6940340995788574,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.386,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 35.33643673296657,
|
| 2676 |
+
"eval_accuracy": 0.9054011321122799,
|
| 2677 |
+
"eval_loss": 0.43167829513549805,
|
| 2678 |
+
"eval_runtime": 56.7945,
|
| 2679 |
+
"eval_samples_per_second": 332.726,
|
| 2680 |
+
"eval_steps_per_second": 10.406,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 35.54803216250529,
|
| 2685 |
+
"grad_norm": 1.6332502365112305,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.3868,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 35.54803216250529,
|
| 2692 |
+
"eval_accuracy": 0.9057828270899271,
|
| 2693 |
+
"eval_loss": 0.4301559627056122,
|
| 2694 |
+
"eval_runtime": 56.7482,
|
| 2695 |
+
"eval_samples_per_second": 332.997,
|
| 2696 |
+
"eval_steps_per_second": 10.414,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 35.75962759204401,
|
| 2701 |
+
"grad_norm": 1.5488901138305664,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.3895,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 35.75962759204401,
|
| 2708 |
+
"eval_accuracy": 0.9057984295149871,
|
| 2709 |
+
"eval_loss": 0.43114641308784485,
|
| 2710 |
+
"eval_runtime": 56.9082,
|
| 2711 |
+
"eval_samples_per_second": 332.061,
|
| 2712 |
+
"eval_steps_per_second": 10.385,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 35.97122302158273,
|
| 2717 |
+
"grad_norm": 1.74102783203125,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.3835,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 35.97122302158273,
|
| 2724 |
+
"eval_accuracy": 0.9061342667345084,
|
| 2725 |
+
"eval_loss": 0.4316621720790863,
|
| 2726 |
+
"eval_runtime": 56.8198,
|
| 2727 |
+
"eval_samples_per_second": 332.578,
|
| 2728 |
+
"eval_steps_per_second": 10.401,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 36.182818451121456,
|
| 2733 |
+
"grad_norm": 1.9937413930892944,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.3866,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 36.182818451121456,
|
| 2740 |
+
"eval_accuracy": 0.9058441609643859,
|
| 2741 |
+
"eval_loss": 0.43131837248802185,
|
| 2742 |
+
"eval_runtime": 56.8234,
|
| 2743 |
+
"eval_samples_per_second": 332.557,
|
| 2744 |
+
"eval_steps_per_second": 10.401,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 36.394413880660174,
|
| 2749 |
+
"grad_norm": 1.7854551076889038,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.3782,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 36.394413880660174,
|
| 2756 |
+
"eval_accuracy": 0.9063865664190446,
|
| 2757 |
+
"eval_loss": 0.4299083948135376,
|
| 2758 |
+
"eval_runtime": 56.7713,
|
| 2759 |
+
"eval_samples_per_second": 332.862,
|
| 2760 |
+
"eval_steps_per_second": 10.41,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 36.6060093101989,
|
| 2765 |
+
"grad_norm": 1.811729907989502,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.3824,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 36.6060093101989,
|
| 2772 |
+
"eval_accuracy": 0.9066919797149291,
|
| 2773 |
+
"eval_loss": 0.42806556820869446,
|
| 2774 |
+
"eval_runtime": 56.8409,
|
| 2775 |
+
"eval_samples_per_second": 332.454,
|
| 2776 |
+
"eval_steps_per_second": 10.397,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 36.817604739737625,
|
| 2781 |
+
"grad_norm": 1.794323205947876,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.3809,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 36.817604739737625,
|
| 2788 |
+
"eval_accuracy": 0.9066849391543508,
|
| 2789 |
+
"eval_loss": 0.4274926483631134,
|
| 2790 |
+
"eval_runtime": 56.7903,
|
| 2791 |
+
"eval_samples_per_second": 332.75,
|
| 2792 |
+
"eval_steps_per_second": 10.407,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 37.02920016927634,
|
| 2797 |
+
"grad_norm": 1.833765983581543,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.3816,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 37.02920016927634,
|
| 2804 |
+
"eval_accuracy": 0.9067738678587011,
|
| 2805 |
+
"eval_loss": 0.42850610613822937,
|
| 2806 |
+
"eval_runtime": 56.8173,
|
| 2807 |
+
"eval_samples_per_second": 332.592,
|
| 2808 |
+
"eval_steps_per_second": 10.402,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 37.24079559881507,
|
| 2813 |
+
"grad_norm": 1.4756935834884644,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.3748,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 37.24079559881507,
|
| 2820 |
+
"eval_accuracy": 0.9064180322306237,
|
| 2821 |
+
"eval_loss": 0.4284866154193878,
|
| 2822 |
+
"eval_runtime": 56.7602,
|
| 2823 |
+
"eval_samples_per_second": 332.927,
|
| 2824 |
+
"eval_steps_per_second": 10.412,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 37.45239102835379,
|
| 2829 |
+
"grad_norm": 1.8428268432617188,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.3777,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 37.45239102835379,
|
| 2836 |
+
"eval_accuracy": 0.9065104011022369,
|
| 2837 |
+
"eval_loss": 0.428924560546875,
|
| 2838 |
+
"eval_runtime": 56.7924,
|
| 2839 |
+
"eval_samples_per_second": 332.738,
|
| 2840 |
+
"eval_steps_per_second": 10.406,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 37.66398645789251,
|
| 2845 |
+
"grad_norm": 1.6545451879501343,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.3803,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 37.66398645789251,
|
| 2852 |
+
"eval_accuracy": 0.9070098926985761,
|
| 2853 |
+
"eval_loss": 0.4258217215538025,
|
| 2854 |
+
"eval_runtime": 56.7499,
|
| 2855 |
+
"eval_samples_per_second": 332.987,
|
| 2856 |
+
"eval_steps_per_second": 10.414,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 37.87558188743123,
|
| 2861 |
+
"grad_norm": 1.4074724912643433,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.3765,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 37.87558188743123,
|
| 2868 |
+
"eval_accuracy": 0.9070845757199019,
|
| 2869 |
+
"eval_loss": 0.4270836114883423,
|
| 2870 |
+
"eval_runtime": 56.7862,
|
| 2871 |
+
"eval_samples_per_second": 332.774,
|
| 2872 |
+
"eval_steps_per_second": 10.407,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 38.08717731696996,
|
| 2877 |
+
"grad_norm": 1.6752023696899414,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.3773,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 38.08717731696996,
|
| 2884 |
+
"eval_accuracy": 0.9067767506632135,
|
| 2885 |
+
"eval_loss": 0.426782488822937,
|
| 2886 |
+
"eval_runtime": 56.924,
|
| 2887 |
+
"eval_samples_per_second": 331.969,
|
| 2888 |
+
"eval_steps_per_second": 10.382,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 38.298772746508675,
|
| 2893 |
+
"grad_norm": 1.8480374813079834,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.3771,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 38.298772746508675,
|
| 2900 |
+
"eval_accuracy": 0.9071141309699655,
|
| 2901 |
+
"eval_loss": 0.4255100190639496,
|
| 2902 |
+
"eval_runtime": 57.0102,
|
| 2903 |
+
"eval_samples_per_second": 331.467,
|
| 2904 |
+
"eval_steps_per_second": 10.367,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 38.5103681760474,
|
| 2909 |
+
"grad_norm": 1.9985958337783813,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.3706,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 38.5103681760474,
|
| 2916 |
+
"eval_accuracy": 0.9076251641060213,
|
| 2917 |
+
"eval_loss": 0.4240247309207916,
|
| 2918 |
+
"eval_runtime": 56.894,
|
| 2919 |
+
"eval_samples_per_second": 332.144,
|
| 2920 |
+
"eval_steps_per_second": 10.388,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 38.72196360558612,
|
| 2925 |
+
"grad_norm": 1.5752030611038208,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.3734,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 38.72196360558612,
|
| 2932 |
+
"eval_accuracy": 0.9080514414097821,
|
| 2933 |
+
"eval_loss": 0.42054280638694763,
|
| 2934 |
+
"eval_runtime": 56.8695,
|
| 2935 |
+
"eval_samples_per_second": 332.287,
|
| 2936 |
+
"eval_steps_per_second": 10.392,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 38.933559035124844,
|
| 2941 |
+
"grad_norm": 1.8178997039794922,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.3707,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 38.933559035124844,
|
| 2948 |
+
"eval_accuracy": 0.9077427484564671,
|
| 2949 |
+
"eval_loss": 0.4216785132884979,
|
| 2950 |
+
"eval_runtime": 56.8577,
|
| 2951 |
+
"eval_samples_per_second": 332.356,
|
| 2952 |
+
"eval_steps_per_second": 10.394,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 39.14515446466356,
|
| 2957 |
+
"grad_norm": 1.4444594383239746,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.3702,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 39.14515446466356,
|
| 2964 |
+
"eval_accuracy": 0.9075348793326441,
|
| 2965 |
+
"eval_loss": 0.42147237062454224,
|
| 2966 |
+
"eval_runtime": 57.0156,
|
| 2967 |
+
"eval_samples_per_second": 331.436,
|
| 2968 |
+
"eval_steps_per_second": 10.366,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 39.35674989420229,
|
| 2973 |
+
"grad_norm": 1.6367250680923462,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.3761,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 39.35674989420229,
|
| 2980 |
+
"eval_accuracy": 0.9075161822246184,
|
| 2981 |
+
"eval_loss": 0.424545556306839,
|
| 2982 |
+
"eval_runtime": 56.8731,
|
| 2983 |
+
"eval_samples_per_second": 332.266,
|
| 2984 |
+
"eval_steps_per_second": 10.392,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 39.568345323741006,
|
| 2989 |
+
"grad_norm": 1.672502875328064,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.3706,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 39.568345323741006,
|
| 2996 |
+
"eval_accuracy": 0.9074108582464906,
|
| 2997 |
+
"eval_loss": 0.4237205684185028,
|
| 2998 |
+
"eval_runtime": 56.9521,
|
| 2999 |
+
"eval_samples_per_second": 331.805,
|
| 3000 |
+
"eval_steps_per_second": 10.377,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 39.77994075327973,
|
| 3005 |
+
"grad_norm": 1.776159644126892,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.3716,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 39.77994075327973,
|
| 3012 |
+
"eval_accuracy": 0.9079466305982884,
|
| 3013 |
+
"eval_loss": 0.42235279083251953,
|
| 3014 |
+
"eval_runtime": 56.8969,
|
| 3015 |
+
"eval_samples_per_second": 332.127,
|
| 3016 |
+
"eval_steps_per_second": 10.387,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 39.99153618281845,
|
| 3021 |
+
"grad_norm": 1.527785062789917,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.3678,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 39.99153618281845,
|
| 3028 |
+
"eval_accuracy": 0.9078065496456583,
|
| 3029 |
+
"eval_loss": 0.42145270109176636,
|
| 3030 |
+
"eval_runtime": 56.9082,
|
| 3031 |
+
"eval_samples_per_second": 332.061,
|
| 3032 |
+
"eval_steps_per_second": 10.385,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 40.203131612357176,
|
| 3037 |
+
"grad_norm": 1.7030891180038452,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.3747,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 40.203131612357176,
|
| 3044 |
+
"eval_accuracy": 0.9082217566630962,
|
| 3045 |
+
"eval_loss": 0.42225930094718933,
|
| 3046 |
+
"eval_runtime": 56.9229,
|
| 3047 |
+
"eval_samples_per_second": 331.976,
|
| 3048 |
+
"eval_steps_per_second": 10.382,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 40.414727041895894,
|
| 3053 |
+
"grad_norm": 1.7826294898986816,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.3737,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 40.414727041895894,
|
| 3060 |
+
"eval_accuracy": 0.9080339951390424,
|
| 3061 |
+
"eval_loss": 0.4210836589336395,
|
| 3062 |
+
"eval_runtime": 56.9513,
|
| 3063 |
+
"eval_samples_per_second": 331.81,
|
| 3064 |
+
"eval_steps_per_second": 10.377,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 40.62632247143462,
|
| 3069 |
+
"grad_norm": 1.5348705053329468,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.3679,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 40.62632247143462,
|
| 3076 |
+
"eval_accuracy": 0.9078710265315678,
|
| 3077 |
+
"eval_loss": 0.42169785499572754,
|
| 3078 |
+
"eval_runtime": 56.9465,
|
| 3079 |
+
"eval_samples_per_second": 331.838,
|
| 3080 |
+
"eval_steps_per_second": 10.378,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 40.83791790097334,
|
| 3085 |
+
"grad_norm": 1.6769896745681763,
|
| 3086 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 3087 |
+
"loss": 0.3642,
|
| 3088 |
+
"step": 96500
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 40.83791790097334,
|
| 3092 |
+
"eval_accuracy": 0.9081198632994205,
|
| 3093 |
+
"eval_loss": 0.42042461037635803,
|
| 3094 |
+
"eval_runtime": 56.8975,
|
| 3095 |
+
"eval_samples_per_second": 332.124,
|
| 3096 |
+
"eval_steps_per_second": 10.387,
|
| 3097 |
+
"step": 96500
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 41.04951333051206,
|
| 3101 |
+
"grad_norm": 1.506687879562378,
|
| 3102 |
+
"learning_rate": 3e-06,
|
| 3103 |
+
"loss": 0.3646,
|
| 3104 |
+
"step": 97000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 41.04951333051206,
|
| 3108 |
+
"eval_accuracy": 0.9083935195759528,
|
| 3109 |
+
"eval_loss": 0.4203515648841858,
|
| 3110 |
+
"eval_runtime": 56.9549,
|
| 3111 |
+
"eval_samples_per_second": 331.789,
|
| 3112 |
+
"eval_steps_per_second": 10.377,
|
| 3113 |
+
"step": 97000
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 41.26110876005078,
|
| 3117 |
+
"grad_norm": 1.5605355501174927,
|
| 3118 |
+
"learning_rate": 2.5e-06,
|
| 3119 |
+
"loss": 0.3663,
|
| 3120 |
+
"step": 97500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 41.26110876005078,
|
| 3124 |
+
"eval_accuracy": 0.9083660447561517,
|
| 3125 |
+
"eval_loss": 0.41981303691864014,
|
| 3126 |
+
"eval_runtime": 56.9268,
|
| 3127 |
+
"eval_samples_per_second": 331.953,
|
| 3128 |
+
"eval_steps_per_second": 10.382,
|
| 3129 |
+
"step": 97500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 41.47270418958951,
|
| 3133 |
+
"grad_norm": 1.7838058471679688,
|
| 3134 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 3135 |
+
"loss": 0.3719,
|
| 3136 |
+
"step": 98000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 41.47270418958951,
|
| 3140 |
+
"eval_accuracy": 0.9090116466984204,
|
| 3141 |
+
"eval_loss": 0.4177597463130951,
|
| 3142 |
+
"eval_runtime": 56.8615,
|
| 3143 |
+
"eval_samples_per_second": 332.334,
|
| 3144 |
+
"eval_steps_per_second": 10.394,
|
| 3145 |
+
"step": 98000
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 41.684299619128225,
|
| 3149 |
+
"grad_norm": 1.6036288738250732,
|
| 3150 |
+
"learning_rate": 1.5e-06,
|
| 3151 |
+
"loss": 0.3663,
|
| 3152 |
+
"step": 98500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 41.684299619128225,
|
| 3156 |
+
"eval_accuracy": 0.9089590262822165,
|
| 3157 |
+
"eval_loss": 0.4168604016304016,
|
| 3158 |
+
"eval_runtime": 56.9476,
|
| 3159 |
+
"eval_samples_per_second": 331.831,
|
| 3160 |
+
"eval_steps_per_second": 10.378,
|
| 3161 |
+
"step": 98500
|
| 3162 |
+
}
|
| 3163 |
+
],
|
| 3164 |
+
"logging_steps": 500,
|
| 3165 |
+
"max_steps": 100000,
|
| 3166 |
+
"num_input_tokens_seen": 0,
|
| 3167 |
+
"num_train_epochs": 43,
|
| 3168 |
+
"save_steps": 500,
|
| 3169 |
+
"stateful_callbacks": {
|
| 3170 |
+
"TrainerControl": {
|
| 3171 |
+
"args": {
|
| 3172 |
+
"should_epoch_stop": false,
|
| 3173 |
+
"should_evaluate": false,
|
| 3174 |
+
"should_log": false,
|
| 3175 |
+
"should_save": true,
|
| 3176 |
+
"should_training_stop": false
|
| 3177 |
+
},
|
| 3178 |
+
"attributes": {}
|
| 3179 |
+
}
|
| 3180 |
+
},
|
| 3181 |
+
"total_flos": 8.30483576347689e+17,
|
| 3182 |
+
"train_batch_size": 16,
|
| 3183 |
+
"trial_name": null,
|
| 3184 |
+
"trial_params": null
|
| 3185 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5818128b17da2198b1a8e5bcd6295a8ad959e513f5068b532394bb02f8e28721
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|