Uploading checkpoint-98500 for mbert - dan-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: dan-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 |
+
# dan-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: 0.8587
|
| 22 |
+
- Accuracy: 0.8314
|
| 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 sha256:79154287316b17931e036caf7e0f1635bb31fa01eb7bbc176c4a5ca1d4095930
|
| 3 |
+
size 711921812
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef69dac806b386b29c83cad902ef57c7b1e883fde3a5e0fa86856a93fcd9a23c
|
| 3 |
+
size 1423965114
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f47184ea54687edf3e39d0b33ef40829288b7eaaaa94d64cb04c5026c1ecd46
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e4865d036cb39d59fcb9d85e66b1cd332b371a459551ea3835f0bbd2f80971c3
|
| 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": 0.8528650403022766,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/mbert/dan-Latn/checkpoint-98500",
|
| 4 |
+
"epoch": 10.286131996658312,
|
| 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": 0.052213868003341685,
|
| 13 |
+
"grad_norm": 4.64046049118042,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.7666,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052213868003341685,
|
| 20 |
+
"eval_accuracy": 0.6961985245587479,
|
| 21 |
+
"eval_loss": 1.741060733795166,
|
| 22 |
+
"eval_runtime": 361.3155,
|
| 23 |
+
"eval_samples_per_second": 175.589,
|
| 24 |
+
"eval_steps_per_second": 5.488,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.10442773600668337,
|
| 29 |
+
"grad_norm": 3.5290920734405518,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.6122,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.10442773600668337,
|
| 36 |
+
"eval_accuracy": 0.7084748820521898,
|
| 37 |
+
"eval_loss": 1.6247525215148926,
|
| 38 |
+
"eval_runtime": 360.2132,
|
| 39 |
+
"eval_samples_per_second": 176.126,
|
| 40 |
+
"eval_steps_per_second": 5.505,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.15664160401002505,
|
| 45 |
+
"grad_norm": 3.951317071914673,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.5487,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.15664160401002505,
|
| 52 |
+
"eval_accuracy": 0.718550082660156,
|
| 53 |
+
"eval_loss": 1.553884506225586,
|
| 54 |
+
"eval_runtime": 361.2942,
|
| 55 |
+
"eval_samples_per_second": 175.599,
|
| 56 |
+
"eval_steps_per_second": 5.489,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.20885547201336674,
|
| 61 |
+
"grad_norm": 3.799274444580078,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 1.4946,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.20885547201336674,
|
| 68 |
+
"eval_accuracy": 0.723134633812131,
|
| 69 |
+
"eval_loss": 1.5116215944290161,
|
| 70 |
+
"eval_runtime": 360.3813,
|
| 71 |
+
"eval_samples_per_second": 176.044,
|
| 72 |
+
"eval_steps_per_second": 5.503,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.26106934001670845,
|
| 77 |
+
"grad_norm": 3.499929428100586,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 1.4573,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.26106934001670845,
|
| 84 |
+
"eval_accuracy": 0.7287729394802086,
|
| 85 |
+
"eval_loss": 1.4748759269714355,
|
| 86 |
+
"eval_runtime": 362.14,
|
| 87 |
+
"eval_samples_per_second": 175.189,
|
| 88 |
+
"eval_steps_per_second": 5.476,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.3132832080200501,
|
| 93 |
+
"grad_norm": 3.6163783073425293,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 1.4332,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.3132832080200501,
|
| 100 |
+
"eval_accuracy": 0.7334675822291515,
|
| 101 |
+
"eval_loss": 1.4382221698760986,
|
| 102 |
+
"eval_runtime": 362.0967,
|
| 103 |
+
"eval_samples_per_second": 175.21,
|
| 104 |
+
"eval_steps_per_second": 5.476,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.3654970760233918,
|
| 109 |
+
"grad_norm": 3.382498264312744,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 1.3914,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.3654970760233918,
|
| 116 |
+
"eval_accuracy": 0.7367988898749408,
|
| 117 |
+
"eval_loss": 1.4078316688537598,
|
| 118 |
+
"eval_runtime": 362.0182,
|
| 119 |
+
"eval_samples_per_second": 175.248,
|
| 120 |
+
"eval_steps_per_second": 5.478,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4177109440267335,
|
| 125 |
+
"grad_norm": 3.2594001293182373,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 1.3843,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4177109440267335,
|
| 132 |
+
"eval_accuracy": 0.7404781561454725,
|
| 133 |
+
"eval_loss": 1.37874174118042,
|
| 134 |
+
"eval_runtime": 362.2257,
|
| 135 |
+
"eval_samples_per_second": 175.148,
|
| 136 |
+
"eval_steps_per_second": 5.474,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.4699248120300752,
|
| 141 |
+
"grad_norm": 3.717684268951416,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 1.3628,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.4699248120300752,
|
| 148 |
+
"eval_accuracy": 0.7430362394212798,
|
| 149 |
+
"eval_loss": 1.373829960823059,
|
| 150 |
+
"eval_runtime": 362.0885,
|
| 151 |
+
"eval_samples_per_second": 175.214,
|
| 152 |
+
"eval_steps_per_second": 5.477,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.5221386800334169,
|
| 157 |
+
"grad_norm": 3.67509126663208,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 1.356,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.5221386800334169,
|
| 164 |
+
"eval_accuracy": 0.7465132147816155,
|
| 165 |
+
"eval_loss": 1.343856930732727,
|
| 166 |
+
"eval_runtime": 362.013,
|
| 167 |
+
"eval_samples_per_second": 175.251,
|
| 168 |
+
"eval_steps_per_second": 5.478,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.5743525480367586,
|
| 173 |
+
"grad_norm": 2.995349407196045,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 1.3275,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.5743525480367586,
|
| 180 |
+
"eval_accuracy": 0.747906382026035,
|
| 181 |
+
"eval_loss": 1.3185055255889893,
|
| 182 |
+
"eval_runtime": 363.6082,
|
| 183 |
+
"eval_samples_per_second": 174.482,
|
| 184 |
+
"eval_steps_per_second": 5.454,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.6265664160401002,
|
| 189 |
+
"grad_norm": 3.0733938217163086,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 1.3192,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.6265664160401002,
|
| 196 |
+
"eval_accuracy": 0.7498967127579244,
|
| 197 |
+
"eval_loss": 1.3267806768417358,
|
| 198 |
+
"eval_runtime": 362.4596,
|
| 199 |
+
"eval_samples_per_second": 175.035,
|
| 200 |
+
"eval_steps_per_second": 5.471,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.6787802840434419,
|
| 205 |
+
"grad_norm": 3.0954840183258057,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 1.3123,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.6787802840434419,
|
| 212 |
+
"eval_accuracy": 0.7517548935891929,
|
| 213 |
+
"eval_loss": 1.3006523847579956,
|
| 214 |
+
"eval_runtime": 363.2093,
|
| 215 |
+
"eval_samples_per_second": 174.673,
|
| 216 |
+
"eval_steps_per_second": 5.46,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.7309941520467836,
|
| 221 |
+
"grad_norm": 3.1289381980895996,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 1.2942,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.7309941520467836,
|
| 228 |
+
"eval_accuracy": 0.7542296938604579,
|
| 229 |
+
"eval_loss": 1.2929292917251587,
|
| 230 |
+
"eval_runtime": 362.9614,
|
| 231 |
+
"eval_samples_per_second": 174.793,
|
| 232 |
+
"eval_steps_per_second": 5.463,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7832080200501254,
|
| 237 |
+
"grad_norm": 3.440656900405884,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 1.2812,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7832080200501254,
|
| 244 |
+
"eval_accuracy": 0.756235706841684,
|
| 245 |
+
"eval_loss": 1.2668538093566895,
|
| 246 |
+
"eval_runtime": 362.1254,
|
| 247 |
+
"eval_samples_per_second": 175.196,
|
| 248 |
+
"eval_steps_per_second": 5.476,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.835421888053467,
|
| 253 |
+
"grad_norm": 4.0889458656311035,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 1.2708,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.835421888053467,
|
| 260 |
+
"eval_accuracy": 0.7571290622439745,
|
| 261 |
+
"eval_loss": 1.283825159072876,
|
| 262 |
+
"eval_runtime": 362.2084,
|
| 263 |
+
"eval_samples_per_second": 175.156,
|
| 264 |
+
"eval_steps_per_second": 5.475,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.8876357560568087,
|
| 269 |
+
"grad_norm": 2.77811598777771,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 1.2759,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.8876357560568087,
|
| 276 |
+
"eval_accuracy": 0.7586107082301347,
|
| 277 |
+
"eval_loss": 1.2615000009536743,
|
| 278 |
+
"eval_runtime": 362.3765,
|
| 279 |
+
"eval_samples_per_second": 175.075,
|
| 280 |
+
"eval_steps_per_second": 5.472,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.9398496240601504,
|
| 285 |
+
"grad_norm": 3.350158929824829,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 1.257,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.9398496240601504,
|
| 292 |
+
"eval_accuracy": 0.7595859575845869,
|
| 293 |
+
"eval_loss": 1.250738501548767,
|
| 294 |
+
"eval_runtime": 362.5674,
|
| 295 |
+
"eval_samples_per_second": 174.983,
|
| 296 |
+
"eval_steps_per_second": 5.469,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.9920634920634921,
|
| 301 |
+
"grad_norm": 3.2566370964050293,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 1.2437,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.9920634920634921,
|
| 308 |
+
"eval_accuracy": 0.7619539601093824,
|
| 309 |
+
"eval_loss": 1.248534083366394,
|
| 310 |
+
"eval_runtime": 362.5746,
|
| 311 |
+
"eval_samples_per_second": 174.979,
|
| 312 |
+
"eval_steps_per_second": 5.469,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.0442773600668338,
|
| 317 |
+
"grad_norm": 3.279320001602173,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 1.2333,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.0442773600668338,
|
| 324 |
+
"eval_accuracy": 0.762933591347759,
|
| 325 |
+
"eval_loss": 1.2348397970199585,
|
| 326 |
+
"eval_runtime": 362.8313,
|
| 327 |
+
"eval_samples_per_second": 174.855,
|
| 328 |
+
"eval_steps_per_second": 5.465,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.0964912280701755,
|
| 333 |
+
"grad_norm": 3.002204418182373,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 1.2204,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.0964912280701755,
|
| 340 |
+
"eval_accuracy": 0.7645752968362766,
|
| 341 |
+
"eval_loss": 1.2361088991165161,
|
| 342 |
+
"eval_runtime": 362.7427,
|
| 343 |
+
"eval_samples_per_second": 174.898,
|
| 344 |
+
"eval_steps_per_second": 5.467,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.1487050960735172,
|
| 349 |
+
"grad_norm": 4.058037281036377,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 1.2116,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.1487050960735172,
|
| 356 |
+
"eval_accuracy": 0.7654839256801539,
|
| 357 |
+
"eval_loss": 1.2149455547332764,
|
| 358 |
+
"eval_runtime": 362.0743,
|
| 359 |
+
"eval_samples_per_second": 175.221,
|
| 360 |
+
"eval_steps_per_second": 5.477,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2009189640768587,
|
| 365 |
+
"grad_norm": 3.364161968231201,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 1.2083,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.2009189640768587,
|
| 372 |
+
"eval_accuracy": 0.7664035053683952,
|
| 373 |
+
"eval_loss": 1.2015998363494873,
|
| 374 |
+
"eval_runtime": 362.482,
|
| 375 |
+
"eval_samples_per_second": 175.024,
|
| 376 |
+
"eval_steps_per_second": 5.471,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.2531328320802004,
|
| 381 |
+
"grad_norm": 2.995591402053833,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 1.2056,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.2531328320802004,
|
| 388 |
+
"eval_accuracy": 0.76765140727518,
|
| 389 |
+
"eval_loss": 1.1961411237716675,
|
| 390 |
+
"eval_runtime": 363.557,
|
| 391 |
+
"eval_samples_per_second": 174.506,
|
| 392 |
+
"eval_steps_per_second": 5.454,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.3053467000835421,
|
| 397 |
+
"grad_norm": 3.010730028152466,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 1.1917,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 1.3053467000835421,
|
| 404 |
+
"eval_accuracy": 0.7685342108740758,
|
| 405 |
+
"eval_loss": 1.1992570161819458,
|
| 406 |
+
"eval_runtime": 362.5156,
|
| 407 |
+
"eval_samples_per_second": 175.008,
|
| 408 |
+
"eval_steps_per_second": 5.47,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.3575605680868839,
|
| 413 |
+
"grad_norm": 3.052395820617676,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 1.1905,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.3575605680868839,
|
| 420 |
+
"eval_accuracy": 0.7699597470201401,
|
| 421 |
+
"eval_loss": 1.2011651992797852,
|
| 422 |
+
"eval_runtime": 361.326,
|
| 423 |
+
"eval_samples_per_second": 175.584,
|
| 424 |
+
"eval_steps_per_second": 5.488,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.4097744360902256,
|
| 429 |
+
"grad_norm": 2.883955717086792,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 1.1839,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.4097744360902256,
|
| 436 |
+
"eval_accuracy": 0.7708384570352297,
|
| 437 |
+
"eval_loss": 1.1753507852554321,
|
| 438 |
+
"eval_runtime": 362.9321,
|
| 439 |
+
"eval_samples_per_second": 174.807,
|
| 440 |
+
"eval_steps_per_second": 5.464,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.4619883040935673,
|
| 445 |
+
"grad_norm": 2.7026546001434326,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 1.1724,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.4619883040935673,
|
| 452 |
+
"eval_accuracy": 0.7709277011489558,
|
| 453 |
+
"eval_loss": 1.168317198753357,
|
| 454 |
+
"eval_runtime": 362.8518,
|
| 455 |
+
"eval_samples_per_second": 174.845,
|
| 456 |
+
"eval_steps_per_second": 5.465,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.514202172096909,
|
| 461 |
+
"grad_norm": 2.8576297760009766,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 1.1758,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.514202172096909,
|
| 468 |
+
"eval_accuracy": 0.7720082438542706,
|
| 469 |
+
"eval_loss": 1.1711252927780151,
|
| 470 |
+
"eval_runtime": 362.6486,
|
| 471 |
+
"eval_samples_per_second": 174.943,
|
| 472 |
+
"eval_steps_per_second": 5.468,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.5664160401002505,
|
| 477 |
+
"grad_norm": 3.314800500869751,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 1.1656,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.5664160401002505,
|
| 484 |
+
"eval_accuracy": 0.7731014057112545,
|
| 485 |
+
"eval_loss": NaN,
|
| 486 |
+
"eval_runtime": 362.4489,
|
| 487 |
+
"eval_samples_per_second": 175.04,
|
| 488 |
+
"eval_steps_per_second": 5.471,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.6186299081035922,
|
| 493 |
+
"grad_norm": 2.740029811859131,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 1.1659,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.6186299081035922,
|
| 500 |
+
"eval_accuracy": 0.7745773535784676,
|
| 501 |
+
"eval_loss": 1.1655621528625488,
|
| 502 |
+
"eval_runtime": 362.0759,
|
| 503 |
+
"eval_samples_per_second": 175.22,
|
| 504 |
+
"eval_steps_per_second": 5.477,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.670843776106934,
|
| 509 |
+
"grad_norm": 2.9377198219299316,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 1.1512,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.670843776106934,
|
| 516 |
+
"eval_accuracy": 0.7745256599790964,
|
| 517 |
+
"eval_loss": 1.1678109169006348,
|
| 518 |
+
"eval_runtime": 362.3938,
|
| 519 |
+
"eval_samples_per_second": 175.066,
|
| 520 |
+
"eval_steps_per_second": 5.472,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.7230576441102756,
|
| 525 |
+
"grad_norm": 3.2804691791534424,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 1.1394,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.7230576441102756,
|
| 532 |
+
"eval_accuracy": 0.7751363378115868,
|
| 533 |
+
"eval_loss": 1.1540894508361816,
|
| 534 |
+
"eval_runtime": 362.0672,
|
| 535 |
+
"eval_samples_per_second": 175.224,
|
| 536 |
+
"eval_steps_per_second": 5.477,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.7752715121136173,
|
| 541 |
+
"grad_norm": 3.486011028289795,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 1.1382,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.7752715121136173,
|
| 548 |
+
"eval_accuracy": 0.7765838022142181,
|
| 549 |
+
"eval_loss": 1.1509926319122314,
|
| 550 |
+
"eval_runtime": 362.2521,
|
| 551 |
+
"eval_samples_per_second": 175.135,
|
| 552 |
+
"eval_steps_per_second": 5.474,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.827485380116959,
|
| 557 |
+
"grad_norm": 2.56192684173584,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 1.1448,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.827485380116959,
|
| 564 |
+
"eval_accuracy": 0.7774433499499322,
|
| 565 |
+
"eval_loss": 1.150856375694275,
|
| 566 |
+
"eval_runtime": 361.437,
|
| 567 |
+
"eval_samples_per_second": 175.53,
|
| 568 |
+
"eval_steps_per_second": 5.486,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.8796992481203008,
|
| 573 |
+
"grad_norm": 2.9113447666168213,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 1.1376,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.8796992481203008,
|
| 580 |
+
"eval_accuracy": 0.7779668095978695,
|
| 581 |
+
"eval_loss": 1.1359950304031372,
|
| 582 |
+
"eval_runtime": 362.1516,
|
| 583 |
+
"eval_samples_per_second": 175.184,
|
| 584 |
+
"eval_steps_per_second": 5.476,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.9319131161236425,
|
| 589 |
+
"grad_norm": 2.8685195446014404,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 1.1306,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.9319131161236425,
|
| 596 |
+
"eval_accuracy": 0.7786356827454954,
|
| 597 |
+
"eval_loss": 1.1356443166732788,
|
| 598 |
+
"eval_runtime": 367.0737,
|
| 599 |
+
"eval_samples_per_second": 172.835,
|
| 600 |
+
"eval_steps_per_second": 5.402,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.9841269841269842,
|
| 605 |
+
"grad_norm": 3.4618313312530518,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 1.1379,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.9841269841269842,
|
| 612 |
+
"eval_accuracy": 0.7796485696787286,
|
| 613 |
+
"eval_loss": 1.1357932090759277,
|
| 614 |
+
"eval_runtime": 367.2471,
|
| 615 |
+
"eval_samples_per_second": 172.753,
|
| 616 |
+
"eval_steps_per_second": 5.4,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 2.036340852130326,
|
| 621 |
+
"grad_norm": 3.1010448932647705,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 1.109,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.036340852130326,
|
| 628 |
+
"eval_accuracy": 0.7807163576099158,
|
| 629 |
+
"eval_loss": 1.1260879039764404,
|
| 630 |
+
"eval_runtime": 367.0489,
|
| 631 |
+
"eval_samples_per_second": 172.846,
|
| 632 |
+
"eval_steps_per_second": 5.403,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.0885547201336676,
|
| 637 |
+
"grad_norm": 3.1995084285736084,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 1.1047,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.0885547201336676,
|
| 644 |
+
"eval_accuracy": 0.7806008208598338,
|
| 645 |
+
"eval_loss": 1.1227824687957764,
|
| 646 |
+
"eval_runtime": 367.4627,
|
| 647 |
+
"eval_samples_per_second": 172.652,
|
| 648 |
+
"eval_steps_per_second": 5.396,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.1407685881370093,
|
| 653 |
+
"grad_norm": 2.741629123687744,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 1.1042,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.1407685881370093,
|
| 660 |
+
"eval_accuracy": 0.7819552122122398,
|
| 661 |
+
"eval_loss": 1.1106891632080078,
|
| 662 |
+
"eval_runtime": 369.2375,
|
| 663 |
+
"eval_samples_per_second": 171.822,
|
| 664 |
+
"eval_steps_per_second": 5.371,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 2.192982456140351,
|
| 669 |
+
"grad_norm": 3.0432229042053223,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 1.1071,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.192982456140351,
|
| 676 |
+
"eval_accuracy": 0.7825761226202902,
|
| 677 |
+
"eval_loss": 1.105894923210144,
|
| 678 |
+
"eval_runtime": 368.3863,
|
| 679 |
+
"eval_samples_per_second": 172.219,
|
| 680 |
+
"eval_steps_per_second": 5.383,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.2451963241436927,
|
| 685 |
+
"grad_norm": 2.60355806350708,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 1.1,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.2451963241436927,
|
| 692 |
+
"eval_accuracy": 0.7824829036453502,
|
| 693 |
+
"eval_loss": 1.1067482233047485,
|
| 694 |
+
"eval_runtime": 369.3294,
|
| 695 |
+
"eval_samples_per_second": 171.779,
|
| 696 |
+
"eval_steps_per_second": 5.369,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.2974101921470345,
|
| 701 |
+
"grad_norm": 3.290499448776245,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 1.094,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.2974101921470345,
|
| 708 |
+
"eval_accuracy": 0.7835772317290801,
|
| 709 |
+
"eval_loss": 1.1064658164978027,
|
| 710 |
+
"eval_runtime": 371.2345,
|
| 711 |
+
"eval_samples_per_second": 170.897,
|
| 712 |
+
"eval_steps_per_second": 5.342,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.3496240601503757,
|
| 717 |
+
"grad_norm": 2.749668598175049,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 1.0868,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.3496240601503757,
|
| 724 |
+
"eval_accuracy": 0.7844280402341692,
|
| 725 |
+
"eval_loss": 1.1136114597320557,
|
| 726 |
+
"eval_runtime": 370.2159,
|
| 727 |
+
"eval_samples_per_second": 171.368,
|
| 728 |
+
"eval_steps_per_second": 5.356,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.4018379281537174,
|
| 733 |
+
"grad_norm": 2.553410768508911,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 1.0938,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.4018379281537174,
|
| 740 |
+
"eval_accuracy": 0.7854044343476029,
|
| 741 |
+
"eval_loss": 1.0965579748153687,
|
| 742 |
+
"eval_runtime": 370.6834,
|
| 743 |
+
"eval_samples_per_second": 171.151,
|
| 744 |
+
"eval_steps_per_second": 5.35,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.454051796157059,
|
| 749 |
+
"grad_norm": 2.7348687648773193,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 1.0881,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.454051796157059,
|
| 756 |
+
"eval_accuracy": 0.7850364727467058,
|
| 757 |
+
"eval_loss": 1.1107020378112793,
|
| 758 |
+
"eval_runtime": 369.6648,
|
| 759 |
+
"eval_samples_per_second": 171.623,
|
| 760 |
+
"eval_steps_per_second": 5.364,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.506265664160401,
|
| 765 |
+
"grad_norm": 2.639407157897949,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 1.0885,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.506265664160401,
|
| 772 |
+
"eval_accuracy": 0.7867807742632953,
|
| 773 |
+
"eval_loss": 1.0890835523605347,
|
| 774 |
+
"eval_runtime": 369.0693,
|
| 775 |
+
"eval_samples_per_second": 171.9,
|
| 776 |
+
"eval_steps_per_second": 5.373,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 2.5584795321637426,
|
| 781 |
+
"grad_norm": 3.1823441982269287,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 1.077,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.5584795321637426,
|
| 788 |
+
"eval_accuracy": 0.7863467916817448,
|
| 789 |
+
"eval_loss": 1.0876108407974243,
|
| 790 |
+
"eval_runtime": 368.1359,
|
| 791 |
+
"eval_samples_per_second": 172.336,
|
| 792 |
+
"eval_steps_per_second": 5.387,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.6106934001670843,
|
| 797 |
+
"grad_norm": 2.6731505393981934,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 1.078,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.6106934001670843,
|
| 804 |
+
"eval_accuracy": 0.7872558672911878,
|
| 805 |
+
"eval_loss": 1.0856846570968628,
|
| 806 |
+
"eval_runtime": 367.5868,
|
| 807 |
+
"eval_samples_per_second": 172.593,
|
| 808 |
+
"eval_steps_per_second": 5.395,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.662907268170426,
|
| 813 |
+
"grad_norm": 2.8799052238464355,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 1.0759,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.662907268170426,
|
| 820 |
+
"eval_accuracy": 0.788363306685935,
|
| 821 |
+
"eval_loss": 1.080370545387268,
|
| 822 |
+
"eval_runtime": 368.0395,
|
| 823 |
+
"eval_samples_per_second": 172.381,
|
| 824 |
+
"eval_steps_per_second": 5.388,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.7151211361737677,
|
| 829 |
+
"grad_norm": 4.7506208419799805,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 1.072,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.7151211361737677,
|
| 836 |
+
"eval_accuracy": 0.7885482778687157,
|
| 837 |
+
"eval_loss": 1.08516526222229,
|
| 838 |
+
"eval_runtime": 366.0953,
|
| 839 |
+
"eval_samples_per_second": 173.296,
|
| 840 |
+
"eval_steps_per_second": 5.417,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.7673350041771094,
|
| 845 |
+
"grad_norm": 2.411212682723999,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 1.0655,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.7673350041771094,
|
| 852 |
+
"eval_accuracy": 0.7889261779231521,
|
| 853 |
+
"eval_loss": 1.0773206949234009,
|
| 854 |
+
"eval_runtime": 366.4448,
|
| 855 |
+
"eval_samples_per_second": 173.131,
|
| 856 |
+
"eval_steps_per_second": 5.411,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 2.819548872180451,
|
| 861 |
+
"grad_norm": 2.6230037212371826,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 1.0747,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 2.819548872180451,
|
| 868 |
+
"eval_accuracy": 0.7891164707308205,
|
| 869 |
+
"eval_loss": 1.0783028602600098,
|
| 870 |
+
"eval_runtime": 367.8248,
|
| 871 |
+
"eval_samples_per_second": 172.482,
|
| 872 |
+
"eval_steps_per_second": 5.391,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 2.871762740183793,
|
| 877 |
+
"grad_norm": 2.9123287200927734,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 1.0691,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 2.871762740183793,
|
| 884 |
+
"eval_accuracy": 0.7904697432780537,
|
| 885 |
+
"eval_loss": 1.0683778524398804,
|
| 886 |
+
"eval_runtime": 365.814,
|
| 887 |
+
"eval_samples_per_second": 173.43,
|
| 888 |
+
"eval_steps_per_second": 5.421,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 2.9239766081871346,
|
| 893 |
+
"grad_norm": 2.716794967651367,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 1.0581,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.9239766081871346,
|
| 900 |
+
"eval_accuracy": 0.7907833141699024,
|
| 901 |
+
"eval_loss": 1.0698238611221313,
|
| 902 |
+
"eval_runtime": 368.4835,
|
| 903 |
+
"eval_samples_per_second": 172.173,
|
| 904 |
+
"eval_steps_per_second": 5.382,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.9761904761904763,
|
| 909 |
+
"grad_norm": 2.900315761566162,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 1.0576,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.9761904761904763,
|
| 916 |
+
"eval_accuracy": 0.7912189924393805,
|
| 917 |
+
"eval_loss": 1.0527312755584717,
|
| 918 |
+
"eval_runtime": 367.3342,
|
| 919 |
+
"eval_samples_per_second": 172.712,
|
| 920 |
+
"eval_steps_per_second": 5.398,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 3.028404344193818,
|
| 925 |
+
"grad_norm": 2.733102321624756,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 1.0491,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 3.028404344193818,
|
| 932 |
+
"eval_accuracy": 0.7912027755966868,
|
| 933 |
+
"eval_loss": 1.0649120807647705,
|
| 934 |
+
"eval_runtime": 367.5821,
|
| 935 |
+
"eval_samples_per_second": 172.595,
|
| 936 |
+
"eval_steps_per_second": 5.395,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 3.0806182121971597,
|
| 941 |
+
"grad_norm": 2.990572452545166,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 1.0429,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 3.0806182121971597,
|
| 948 |
+
"eval_accuracy": 0.791866931892446,
|
| 949 |
+
"eval_loss": 1.065590262413025,
|
| 950 |
+
"eval_runtime": 367.6583,
|
| 951 |
+
"eval_samples_per_second": 172.56,
|
| 952 |
+
"eval_steps_per_second": 5.394,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 3.1328320802005014,
|
| 957 |
+
"grad_norm": 2.6242377758026123,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 1.0416,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 3.1328320802005014,
|
| 964 |
+
"eval_accuracy": 0.793420728495659,
|
| 965 |
+
"eval_loss": 1.056522250175476,
|
| 966 |
+
"eval_runtime": 367.1286,
|
| 967 |
+
"eval_samples_per_second": 172.809,
|
| 968 |
+
"eval_steps_per_second": 5.401,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 3.185045948203843,
|
| 973 |
+
"grad_norm": 2.6089892387390137,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 1.0388,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 3.185045948203843,
|
| 980 |
+
"eval_accuracy": 0.792953820707681,
|
| 981 |
+
"eval_loss": 1.0602775812149048,
|
| 982 |
+
"eval_runtime": 367.5358,
|
| 983 |
+
"eval_samples_per_second": 172.617,
|
| 984 |
+
"eval_steps_per_second": 5.395,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.2372598162071844,
|
| 989 |
+
"grad_norm": 2.6940767765045166,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 1.0351,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.2372598162071844,
|
| 996 |
+
"eval_accuracy": 0.7933434972497504,
|
| 997 |
+
"eval_loss": 1.0466923713684082,
|
| 998 |
+
"eval_runtime": 367.3002,
|
| 999 |
+
"eval_samples_per_second": 172.728,
|
| 1000 |
+
"eval_steps_per_second": 5.399,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 3.2894736842105265,
|
| 1005 |
+
"grad_norm": 2.5024611949920654,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 1.0364,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 3.2894736842105265,
|
| 1012 |
+
"eval_accuracy": 0.7938457165216277,
|
| 1013 |
+
"eval_loss": 1.0515408515930176,
|
| 1014 |
+
"eval_runtime": 367.3309,
|
| 1015 |
+
"eval_samples_per_second": 172.713,
|
| 1016 |
+
"eval_steps_per_second": 5.398,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 3.341687552213868,
|
| 1021 |
+
"grad_norm": 2.594076156616211,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 1.0352,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 3.341687552213868,
|
| 1028 |
+
"eval_accuracy": 0.7949145678709858,
|
| 1029 |
+
"eval_loss": 1.0443620681762695,
|
| 1030 |
+
"eval_runtime": 367.752,
|
| 1031 |
+
"eval_samples_per_second": 172.516,
|
| 1032 |
+
"eval_steps_per_second": 5.392,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 3.3939014202172095,
|
| 1037 |
+
"grad_norm": 4.887377738952637,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 1.0269,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.3939014202172095,
|
| 1044 |
+
"eval_accuracy": 0.7945826018068661,
|
| 1045 |
+
"eval_loss": 1.0488784313201904,
|
| 1046 |
+
"eval_runtime": 368.3885,
|
| 1047 |
+
"eval_samples_per_second": 172.218,
|
| 1048 |
+
"eval_steps_per_second": 5.383,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 3.4461152882205512,
|
| 1053 |
+
"grad_norm": 3.1437246799468994,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 1.0293,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 3.4461152882205512,
|
| 1060 |
+
"eval_accuracy": 0.795481144785727,
|
| 1061 |
+
"eval_loss": 1.0364991426467896,
|
| 1062 |
+
"eval_runtime": 366.6755,
|
| 1063 |
+
"eval_samples_per_second": 173.022,
|
| 1064 |
+
"eval_steps_per_second": 5.408,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 3.498329156223893,
|
| 1069 |
+
"grad_norm": 2.490694761276245,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 1.0179,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 3.498329156223893,
|
| 1076 |
+
"eval_accuracy": 0.7964686907750312,
|
| 1077 |
+
"eval_loss": 1.043065071105957,
|
| 1078 |
+
"eval_runtime": 367.4767,
|
| 1079 |
+
"eval_samples_per_second": 172.645,
|
| 1080 |
+
"eval_steps_per_second": 5.396,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 3.5505430242272347,
|
| 1085 |
+
"grad_norm": 2.9056098461151123,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 1.0147,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 3.5505430242272347,
|
| 1092 |
+
"eval_accuracy": 0.7962407550483477,
|
| 1093 |
+
"eval_loss": NaN,
|
| 1094 |
+
"eval_runtime": 367.3507,
|
| 1095 |
+
"eval_samples_per_second": 172.704,
|
| 1096 |
+
"eval_steps_per_second": 5.398,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 3.6027568922305764,
|
| 1101 |
+
"grad_norm": 2.655351161956787,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 1.0112,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 3.6027568922305764,
|
| 1108 |
+
"eval_accuracy": 0.7970424400295055,
|
| 1109 |
+
"eval_loss": 1.033272624015808,
|
| 1110 |
+
"eval_runtime": 367.4604,
|
| 1111 |
+
"eval_samples_per_second": 172.653,
|
| 1112 |
+
"eval_steps_per_second": 5.397,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 3.654970760233918,
|
| 1117 |
+
"grad_norm": 2.922175168991089,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 1.0145,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 3.654970760233918,
|
| 1124 |
+
"eval_accuracy": 0.7967277004610398,
|
| 1125 |
+
"eval_loss": 1.0407856702804565,
|
| 1126 |
+
"eval_runtime": 367.6311,
|
| 1127 |
+
"eval_samples_per_second": 172.572,
|
| 1128 |
+
"eval_steps_per_second": 5.394,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.70718462823726,
|
| 1133 |
+
"grad_norm": 2.7691986560821533,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 1.0133,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.70718462823726,
|
| 1140 |
+
"eval_accuracy": 0.7978057552609327,
|
| 1141 |
+
"eval_loss": 1.0315524339675903,
|
| 1142 |
+
"eval_runtime": 368.5127,
|
| 1143 |
+
"eval_samples_per_second": 172.16,
|
| 1144 |
+
"eval_steps_per_second": 5.381,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 3.7593984962406015,
|
| 1149 |
+
"grad_norm": 2.718721389770508,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 1.0123,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 3.7593984962406015,
|
| 1156 |
+
"eval_accuracy": 0.7984869599976302,
|
| 1157 |
+
"eval_loss": 1.0284632444381714,
|
| 1158 |
+
"eval_runtime": 367.5592,
|
| 1159 |
+
"eval_samples_per_second": 172.606,
|
| 1160 |
+
"eval_steps_per_second": 5.395,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 3.8116123642439432,
|
| 1165 |
+
"grad_norm": 2.4457788467407227,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 1.008,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 3.8116123642439432,
|
| 1172 |
+
"eval_accuracy": 0.7985738639575599,
|
| 1173 |
+
"eval_loss": 1.0290788412094116,
|
| 1174 |
+
"eval_runtime": 367.5585,
|
| 1175 |
+
"eval_samples_per_second": 172.607,
|
| 1176 |
+
"eval_steps_per_second": 5.395,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 3.863826232247285,
|
| 1181 |
+
"grad_norm": 3.0719549655914307,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.9995,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 3.863826232247285,
|
| 1188 |
+
"eval_accuracy": 0.7990341348986376,
|
| 1189 |
+
"eval_loss": 1.0225682258605957,
|
| 1190 |
+
"eval_runtime": 367.5861,
|
| 1191 |
+
"eval_samples_per_second": 172.594,
|
| 1192 |
+
"eval_steps_per_second": 5.395,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 3.9160401002506267,
|
| 1197 |
+
"grad_norm": 2.502732038497925,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 1.0059,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 3.9160401002506267,
|
| 1204 |
+
"eval_accuracy": 0.7990824767003194,
|
| 1205 |
+
"eval_loss": 1.0225658416748047,
|
| 1206 |
+
"eval_runtime": 368.4364,
|
| 1207 |
+
"eval_samples_per_second": 172.195,
|
| 1208 |
+
"eval_steps_per_second": 5.382,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 3.9682539682539684,
|
| 1213 |
+
"grad_norm": 2.615722894668579,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.9908,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 3.9682539682539684,
|
| 1220 |
+
"eval_accuracy": 0.8004074602540703,
|
| 1221 |
+
"eval_loss": NaN,
|
| 1222 |
+
"eval_runtime": 367.3504,
|
| 1223 |
+
"eval_samples_per_second": 172.704,
|
| 1224 |
+
"eval_steps_per_second": 5.398,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 4.02046783625731,
|
| 1229 |
+
"grad_norm": 2.6223480701446533,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 1.0019,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 4.02046783625731,
|
| 1236 |
+
"eval_accuracy": 0.8002175287186218,
|
| 1237 |
+
"eval_loss": 1.0221267938613892,
|
| 1238 |
+
"eval_runtime": 367.5886,
|
| 1239 |
+
"eval_samples_per_second": 172.592,
|
| 1240 |
+
"eval_steps_per_second": 5.395,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 4.072681704260652,
|
| 1245 |
+
"grad_norm": 7.730719089508057,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.9859,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 4.072681704260652,
|
| 1252 |
+
"eval_accuracy": 0.8004946972174642,
|
| 1253 |
+
"eval_loss": 1.0225615501403809,
|
| 1254 |
+
"eval_runtime": 367.5118,
|
| 1255 |
+
"eval_samples_per_second": 172.628,
|
| 1256 |
+
"eval_steps_per_second": 5.396,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 4.124895572263993,
|
| 1261 |
+
"grad_norm": 2.6854565143585205,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.9899,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 4.124895572263993,
|
| 1268 |
+
"eval_accuracy": 0.8006603707418583,
|
| 1269 |
+
"eval_loss": 1.0182828903198242,
|
| 1270 |
+
"eval_runtime": 367.2006,
|
| 1271 |
+
"eval_samples_per_second": 172.775,
|
| 1272 |
+
"eval_steps_per_second": 5.4,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 4.177109440267335,
|
| 1277 |
+
"grad_norm": 2.871232509613037,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.9769,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 4.177109440267335,
|
| 1284 |
+
"eval_accuracy": 0.8012399052534226,
|
| 1285 |
+
"eval_loss": 1.007629632949829,
|
| 1286 |
+
"eval_runtime": 367.698,
|
| 1287 |
+
"eval_samples_per_second": 172.541,
|
| 1288 |
+
"eval_steps_per_second": 5.393,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 4.2293233082706765,
|
| 1293 |
+
"grad_norm": 2.7055728435516357,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.9862,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 4.2293233082706765,
|
| 1300 |
+
"eval_accuracy": 0.8023664324313652,
|
| 1301 |
+
"eval_loss": 1.0073480606079102,
|
| 1302 |
+
"eval_runtime": 367.5166,
|
| 1303 |
+
"eval_samples_per_second": 172.626,
|
| 1304 |
+
"eval_steps_per_second": 5.396,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 4.281537176274019,
|
| 1309 |
+
"grad_norm": 2.628458261489868,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.9784,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 4.281537176274019,
|
| 1316 |
+
"eval_accuracy": 0.8022199523230344,
|
| 1317 |
+
"eval_loss": 1.0004937648773193,
|
| 1318 |
+
"eval_runtime": 368.0641,
|
| 1319 |
+
"eval_samples_per_second": 172.369,
|
| 1320 |
+
"eval_steps_per_second": 5.388,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 4.33375104427736,
|
| 1325 |
+
"grad_norm": 2.5863466262817383,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.974,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 4.33375104427736,
|
| 1332 |
+
"eval_accuracy": 0.8025649357830319,
|
| 1333 |
+
"eval_loss": 1.0010262727737427,
|
| 1334 |
+
"eval_runtime": 367.7203,
|
| 1335 |
+
"eval_samples_per_second": 172.531,
|
| 1336 |
+
"eval_steps_per_second": 5.393,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 4.385964912280702,
|
| 1341 |
+
"grad_norm": 2.748701333999634,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.9828,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 4.385964912280702,
|
| 1348 |
+
"eval_accuracy": 0.8030370401871146,
|
| 1349 |
+
"eval_loss": 0.998631477355957,
|
| 1350 |
+
"eval_runtime": 367.8751,
|
| 1351 |
+
"eval_samples_per_second": 172.458,
|
| 1352 |
+
"eval_steps_per_second": 5.39,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 4.438178780284043,
|
| 1357 |
+
"grad_norm": 2.7004311084747314,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.9863,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.438178780284043,
|
| 1364 |
+
"eval_accuracy": 0.8032171565331311,
|
| 1365 |
+
"eval_loss": 1.0026530027389526,
|
| 1366 |
+
"eval_runtime": 367.6912,
|
| 1367 |
+
"eval_samples_per_second": 172.544,
|
| 1368 |
+
"eval_steps_per_second": 5.393,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 4.4903926482873855,
|
| 1373 |
+
"grad_norm": 2.4777750968933105,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.9725,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 4.4903926482873855,
|
| 1380 |
+
"eval_accuracy": 0.8036803447902678,
|
| 1381 |
+
"eval_loss": 0.9949015378952026,
|
| 1382 |
+
"eval_runtime": 367.6755,
|
| 1383 |
+
"eval_samples_per_second": 172.552,
|
| 1384 |
+
"eval_steps_per_second": 5.393,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 4.542606516290727,
|
| 1389 |
+
"grad_norm": 2.6259942054748535,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.9685,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 4.542606516290727,
|
| 1396 |
+
"eval_accuracy": 0.8042554828282609,
|
| 1397 |
+
"eval_loss": 0.9936387538909912,
|
| 1398 |
+
"eval_runtime": 368.1053,
|
| 1399 |
+
"eval_samples_per_second": 172.35,
|
| 1400 |
+
"eval_steps_per_second": 5.387,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 4.594820384294069,
|
| 1405 |
+
"grad_norm": 3.246826648712158,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.9736,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 4.594820384294069,
|
| 1412 |
+
"eval_accuracy": 0.8043477668789687,
|
| 1413 |
+
"eval_loss": 0.9914335608482361,
|
| 1414 |
+
"eval_runtime": 367.7482,
|
| 1415 |
+
"eval_samples_per_second": 172.518,
|
| 1416 |
+
"eval_steps_per_second": 5.392,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 4.64703425229741,
|
| 1421 |
+
"grad_norm": 2.689743995666504,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.9695,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 4.64703425229741,
|
| 1428 |
+
"eval_accuracy": 0.8044957424284033,
|
| 1429 |
+
"eval_loss": 1.0012022256851196,
|
| 1430 |
+
"eval_runtime": 367.8322,
|
| 1431 |
+
"eval_samples_per_second": 172.478,
|
| 1432 |
+
"eval_steps_per_second": 5.391,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 4.6992481203007515,
|
| 1437 |
+
"grad_norm": 2.6002440452575684,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.9632,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 4.6992481203007515,
|
| 1444 |
+
"eval_accuracy": 0.8053408897805091,
|
| 1445 |
+
"eval_loss": 0.9931978583335876,
|
| 1446 |
+
"eval_runtime": 367.8417,
|
| 1447 |
+
"eval_samples_per_second": 172.474,
|
| 1448 |
+
"eval_steps_per_second": 5.391,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 4.751461988304094,
|
| 1453 |
+
"grad_norm": 2.720722198486328,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.9683,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 4.751461988304094,
|
| 1460 |
+
"eval_accuracy": 0.8055148927829874,
|
| 1461 |
+
"eval_loss": 0.9913645386695862,
|
| 1462 |
+
"eval_runtime": 367.8141,
|
| 1463 |
+
"eval_samples_per_second": 172.487,
|
| 1464 |
+
"eval_steps_per_second": 5.391,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.803675856307435,
|
| 1469 |
+
"grad_norm": 2.6984803676605225,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.9559,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.803675856307435,
|
| 1476 |
+
"eval_accuracy": 0.8059522006843057,
|
| 1477 |
+
"eval_loss": 0.9891652464866638,
|
| 1478 |
+
"eval_runtime": 368.2454,
|
| 1479 |
+
"eval_samples_per_second": 172.285,
|
| 1480 |
+
"eval_steps_per_second": 5.385,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 4.855889724310777,
|
| 1485 |
+
"grad_norm": 2.4419870376586914,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.9592,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 4.855889724310777,
|
| 1492 |
+
"eval_accuracy": 0.805452993603716,
|
| 1493 |
+
"eval_loss": 0.9957113862037659,
|
| 1494 |
+
"eval_runtime": 368.2591,
|
| 1495 |
+
"eval_samples_per_second": 172.278,
|
| 1496 |
+
"eval_steps_per_second": 5.385,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 4.908103592314118,
|
| 1501 |
+
"grad_norm": 2.915653944015503,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.9593,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 4.908103592314118,
|
| 1508 |
+
"eval_accuracy": 0.8068829182930286,
|
| 1509 |
+
"eval_loss": 0.987449049949646,
|
| 1510 |
+
"eval_runtime": 367.9346,
|
| 1511 |
+
"eval_samples_per_second": 172.43,
|
| 1512 |
+
"eval_steps_per_second": 5.39,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 4.9603174603174605,
|
| 1517 |
+
"grad_norm": 2.3721494674682617,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.9521,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 4.9603174603174605,
|
| 1524 |
+
"eval_accuracy": 0.8072047578358524,
|
| 1525 |
+
"eval_loss": 0.9854400157928467,
|
| 1526 |
+
"eval_runtime": 368.1441,
|
| 1527 |
+
"eval_samples_per_second": 172.332,
|
| 1528 |
+
"eval_steps_per_second": 5.386,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 5.012531328320802,
|
| 1533 |
+
"grad_norm": 2.7081363201141357,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.9516,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 5.012531328320802,
|
| 1540 |
+
"eval_accuracy": 0.8074420340361171,
|
| 1541 |
+
"eval_loss": 0.9737870097160339,
|
| 1542 |
+
"eval_runtime": 367.5696,
|
| 1543 |
+
"eval_samples_per_second": 172.601,
|
| 1544 |
+
"eval_steps_per_second": 5.395,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 5.064745196324144,
|
| 1549 |
+
"grad_norm": 2.834805965423584,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.9361,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 5.064745196324144,
|
| 1556 |
+
"eval_accuracy": 0.807780889243512,
|
| 1557 |
+
"eval_loss": NaN,
|
| 1558 |
+
"eval_runtime": 366.5337,
|
| 1559 |
+
"eval_samples_per_second": 173.089,
|
| 1560 |
+
"eval_steps_per_second": 5.41,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 5.116959064327485,
|
| 1565 |
+
"grad_norm": 2.5488121509552,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.9472,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 5.116959064327485,
|
| 1572 |
+
"eval_accuracy": 0.8083244761680142,
|
| 1573 |
+
"eval_loss": 0.9714484214782715,
|
| 1574 |
+
"eval_runtime": 364.0778,
|
| 1575 |
+
"eval_samples_per_second": 174.257,
|
| 1576 |
+
"eval_steps_per_second": 5.447,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 5.169172932330827,
|
| 1581 |
+
"grad_norm": 2.5738179683685303,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.9387,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 5.169172932330827,
|
| 1588 |
+
"eval_accuracy": 0.8083237627586395,
|
| 1589 |
+
"eval_loss": NaN,
|
| 1590 |
+
"eval_runtime": 366.0302,
|
| 1591 |
+
"eval_samples_per_second": 173.327,
|
| 1592 |
+
"eval_steps_per_second": 5.418,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 5.221386800334169,
|
| 1597 |
+
"grad_norm": 2.578078508377075,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.9405,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 5.221386800334169,
|
| 1604 |
+
"eval_accuracy": 0.8092523232767513,
|
| 1605 |
+
"eval_loss": 0.9723265767097473,
|
| 1606 |
+
"eval_runtime": 365.9623,
|
| 1607 |
+
"eval_samples_per_second": 173.359,
|
| 1608 |
+
"eval_steps_per_second": 5.419,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 5.273600668337511,
|
| 1613 |
+
"grad_norm": 3.3929808139801025,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.9386,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 5.273600668337511,
|
| 1620 |
+
"eval_accuracy": 0.8091146550523858,
|
| 1621 |
+
"eval_loss": 0.9661393761634827,
|
| 1622 |
+
"eval_runtime": 365.1784,
|
| 1623 |
+
"eval_samples_per_second": 173.732,
|
| 1624 |
+
"eval_steps_per_second": 5.43,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 5.325814536340852,
|
| 1629 |
+
"grad_norm": 2.3634896278381348,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.933,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 5.325814536340852,
|
| 1636 |
+
"eval_accuracy": 0.808835748295871,
|
| 1637 |
+
"eval_loss": 0.9813971519470215,
|
| 1638 |
+
"eval_runtime": 381.116,
|
| 1639 |
+
"eval_samples_per_second": 166.466,
|
| 1640 |
+
"eval_steps_per_second": 5.203,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 5.378028404344194,
|
| 1645 |
+
"grad_norm": 2.9042844772338867,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.9295,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 5.378028404344194,
|
| 1652 |
+
"eval_accuracy": 0.8103665776447174,
|
| 1653 |
+
"eval_loss": 0.9754624366760254,
|
| 1654 |
+
"eval_runtime": 396.2254,
|
| 1655 |
+
"eval_samples_per_second": 160.118,
|
| 1656 |
+
"eval_steps_per_second": 5.005,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 5.430242272347535,
|
| 1661 |
+
"grad_norm": 2.7546634674072266,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.9367,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 5.430242272347535,
|
| 1668 |
+
"eval_accuracy": 0.8098531380075188,
|
| 1669 |
+
"eval_loss": 0.9821533560752869,
|
| 1670 |
+
"eval_runtime": 400.7427,
|
| 1671 |
+
"eval_samples_per_second": 158.314,
|
| 1672 |
+
"eval_steps_per_second": 4.948,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 5.482456140350878,
|
| 1677 |
+
"grad_norm": 2.8839492797851562,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.9292,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 5.482456140350878,
|
| 1684 |
+
"eval_accuracy": 0.8099045387069566,
|
| 1685 |
+
"eval_loss": 0.974047064781189,
|
| 1686 |
+
"eval_runtime": 399.7171,
|
| 1687 |
+
"eval_samples_per_second": 158.72,
|
| 1688 |
+
"eval_steps_per_second": 4.961,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 5.534670008354219,
|
| 1693 |
+
"grad_norm": 2.3632848262786865,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.9217,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 5.534670008354219,
|
| 1700 |
+
"eval_accuracy": 0.8100555138007703,
|
| 1701 |
+
"eval_loss": 0.9726797938346863,
|
| 1702 |
+
"eval_runtime": 398.2666,
|
| 1703 |
+
"eval_samples_per_second": 159.298,
|
| 1704 |
+
"eval_steps_per_second": 4.979,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 5.586883876357561,
|
| 1709 |
+
"grad_norm": 3.127169609069824,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.9294,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 5.586883876357561,
|
| 1716 |
+
"eval_accuracy": 0.8108817419721464,
|
| 1717 |
+
"eval_loss": 0.9585431218147278,
|
| 1718 |
+
"eval_runtime": 399.6498,
|
| 1719 |
+
"eval_samples_per_second": 158.746,
|
| 1720 |
+
"eval_steps_per_second": 4.962,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 5.639097744360902,
|
| 1725 |
+
"grad_norm": 2.53694748878479,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.9195,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 5.639097744360902,
|
| 1732 |
+
"eval_accuracy": 0.8119484709549883,
|
| 1733 |
+
"eval_loss": 0.957590639591217,
|
| 1734 |
+
"eval_runtime": 394.1783,
|
| 1735 |
+
"eval_samples_per_second": 160.95,
|
| 1736 |
+
"eval_steps_per_second": 5.031,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 5.6913116123642435,
|
| 1741 |
+
"grad_norm": 2.6188199520111084,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.9233,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 5.6913116123642435,
|
| 1748 |
+
"eval_accuracy": 0.8120400051070349,
|
| 1749 |
+
"eval_loss": 0.9545953273773193,
|
| 1750 |
+
"eval_runtime": 395.9182,
|
| 1751 |
+
"eval_samples_per_second": 160.243,
|
| 1752 |
+
"eval_steps_per_second": 5.009,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 5.743525480367586,
|
| 1757 |
+
"grad_norm": 2.502518653869629,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.9272,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 5.743525480367586,
|
| 1764 |
+
"eval_accuracy": 0.8122659898359591,
|
| 1765 |
+
"eval_loss": 0.95594322681427,
|
| 1766 |
+
"eval_runtime": 396.2822,
|
| 1767 |
+
"eval_samples_per_second": 160.096,
|
| 1768 |
+
"eval_steps_per_second": 5.004,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 5.795739348370927,
|
| 1773 |
+
"grad_norm": 2.527303457260132,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.9163,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 5.795739348370927,
|
| 1780 |
+
"eval_accuracy": 0.8124216614631093,
|
| 1781 |
+
"eval_loss": 0.9586062431335449,
|
| 1782 |
+
"eval_runtime": 402.0755,
|
| 1783 |
+
"eval_samples_per_second": 157.789,
|
| 1784 |
+
"eval_steps_per_second": 4.932,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 5.847953216374269,
|
| 1789 |
+
"grad_norm": 2.5056955814361572,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.9249,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 5.847953216374269,
|
| 1796 |
+
"eval_accuracy": 0.8123958167743631,
|
| 1797 |
+
"eval_loss": 0.9501298069953918,
|
| 1798 |
+
"eval_runtime": 397.3012,
|
| 1799 |
+
"eval_samples_per_second": 159.685,
|
| 1800 |
+
"eval_steps_per_second": 4.991,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 5.90016708437761,
|
| 1805 |
+
"grad_norm": 2.876765727996826,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.9167,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 5.90016708437761,
|
| 1812 |
+
"eval_accuracy": 0.8123634219518265,
|
| 1813 |
+
"eval_loss": 0.958740770816803,
|
| 1814 |
+
"eval_runtime": 393.8274,
|
| 1815 |
+
"eval_samples_per_second": 161.093,
|
| 1816 |
+
"eval_steps_per_second": 5.035,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 5.9523809523809526,
|
| 1821 |
+
"grad_norm": 2.457308769226074,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.9191,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 5.9523809523809526,
|
| 1828 |
+
"eval_accuracy": 0.8134356064347583,
|
| 1829 |
+
"eval_loss": 0.9526042938232422,
|
| 1830 |
+
"eval_runtime": 393.3254,
|
| 1831 |
+
"eval_samples_per_second": 161.299,
|
| 1832 |
+
"eval_steps_per_second": 5.042,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 6.004594820384294,
|
| 1837 |
+
"grad_norm": 2.489978075027466,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.9176,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 6.004594820384294,
|
| 1844 |
+
"eval_accuracy": 0.8137190312134606,
|
| 1845 |
+
"eval_loss": 0.9511122703552246,
|
| 1846 |
+
"eval_runtime": 398.7282,
|
| 1847 |
+
"eval_samples_per_second": 159.113,
|
| 1848 |
+
"eval_steps_per_second": 4.973,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 6.056808688387636,
|
| 1853 |
+
"grad_norm": 2.7550694942474365,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.9034,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 6.056808688387636,
|
| 1860 |
+
"eval_accuracy": 0.8137725525224172,
|
| 1861 |
+
"eval_loss": 0.9519580006599426,
|
| 1862 |
+
"eval_runtime": 397.3745,
|
| 1863 |
+
"eval_samples_per_second": 159.655,
|
| 1864 |
+
"eval_steps_per_second": 4.99,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 6.109022556390977,
|
| 1869 |
+
"grad_norm": 2.820486068725586,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.905,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 6.109022556390977,
|
| 1876 |
+
"eval_accuracy": 0.8139618722177003,
|
| 1877 |
+
"eval_loss": 0.9513248801231384,
|
| 1878 |
+
"eval_runtime": 401.2273,
|
| 1879 |
+
"eval_samples_per_second": 158.122,
|
| 1880 |
+
"eval_steps_per_second": 4.942,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 6.161236424394319,
|
| 1885 |
+
"grad_norm": 2.634119749069214,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.902,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 6.161236424394319,
|
| 1892 |
+
"eval_accuracy": 0.8135963134645056,
|
| 1893 |
+
"eval_loss": 0.9465650916099548,
|
| 1894 |
+
"eval_runtime": 402.0909,
|
| 1895 |
+
"eval_samples_per_second": 157.783,
|
| 1896 |
+
"eval_steps_per_second": 4.932,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 6.213450292397661,
|
| 1901 |
+
"grad_norm": 2.8152124881744385,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.9016,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 6.213450292397661,
|
| 1908 |
+
"eval_accuracy": 0.8144102857826985,
|
| 1909 |
+
"eval_loss": 0.9431130886077881,
|
| 1910 |
+
"eval_runtime": 399.633,
|
| 1911 |
+
"eval_samples_per_second": 158.753,
|
| 1912 |
+
"eval_steps_per_second": 4.962,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 6.265664160401003,
|
| 1917 |
+
"grad_norm": 2.478999137878418,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.9018,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 6.265664160401003,
|
| 1924 |
+
"eval_accuracy": 0.8148281916671716,
|
| 1925 |
+
"eval_loss": 0.9465348124504089,
|
| 1926 |
+
"eval_runtime": 401.047,
|
| 1927 |
+
"eval_samples_per_second": 158.193,
|
| 1928 |
+
"eval_steps_per_second": 4.945,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 6.317878028404344,
|
| 1933 |
+
"grad_norm": 2.4472334384918213,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.8894,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 6.317878028404344,
|
| 1940 |
+
"eval_accuracy": 0.8152884197022962,
|
| 1941 |
+
"eval_loss": 0.9336789846420288,
|
| 1942 |
+
"eval_runtime": 399.7265,
|
| 1943 |
+
"eval_samples_per_second": 158.716,
|
| 1944 |
+
"eval_steps_per_second": 4.961,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 6.370091896407686,
|
| 1949 |
+
"grad_norm": 2.6032795906066895,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.8861,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 6.370091896407686,
|
| 1956 |
+
"eval_accuracy": 0.8158878972595904,
|
| 1957 |
+
"eval_loss": 0.9414376616477966,
|
| 1958 |
+
"eval_runtime": 403.8156,
|
| 1959 |
+
"eval_samples_per_second": 157.109,
|
| 1960 |
+
"eval_steps_per_second": 4.911,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 6.4223057644110275,
|
| 1965 |
+
"grad_norm": 2.5782527923583984,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.8929,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 6.4223057644110275,
|
| 1972 |
+
"eval_accuracy": 0.8158353407843104,
|
| 1973 |
+
"eval_loss": 0.9369956254959106,
|
| 1974 |
+
"eval_runtime": 403.5486,
|
| 1975 |
+
"eval_samples_per_second": 157.213,
|
| 1976 |
+
"eval_steps_per_second": 4.914,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 6.474519632414369,
|
| 1981 |
+
"grad_norm": 2.3801369667053223,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.8869,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 6.474519632414369,
|
| 1988 |
+
"eval_accuracy": 0.8156502533666622,
|
| 1989 |
+
"eval_loss": 0.9334683418273926,
|
| 1990 |
+
"eval_runtime": 401.7901,
|
| 1991 |
+
"eval_samples_per_second": 157.901,
|
| 1992 |
+
"eval_steps_per_second": 4.935,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 6.526733500417711,
|
| 1997 |
+
"grad_norm": 2.44163179397583,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.8933,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 6.526733500417711,
|
| 2004 |
+
"eval_accuracy": 0.8168851117616432,
|
| 2005 |
+
"eval_loss": 0.9306558966636658,
|
| 2006 |
+
"eval_runtime": 397.9786,
|
| 2007 |
+
"eval_samples_per_second": 159.413,
|
| 2008 |
+
"eval_steps_per_second": 4.983,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 6.578947368421053,
|
| 2013 |
+
"grad_norm": 2.2924203872680664,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.8862,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 6.578947368421053,
|
| 2020 |
+
"eval_accuracy": 0.8162968497056878,
|
| 2021 |
+
"eval_loss": 0.9353703856468201,
|
| 2022 |
+
"eval_runtime": 394.7488,
|
| 2023 |
+
"eval_samples_per_second": 160.717,
|
| 2024 |
+
"eval_steps_per_second": 5.023,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 6.631161236424394,
|
| 2029 |
+
"grad_norm": 2.552828550338745,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.8948,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 6.631161236424394,
|
| 2036 |
+
"eval_accuracy": 0.8177748484660532,
|
| 2037 |
+
"eval_loss": 0.9219902157783508,
|
| 2038 |
+
"eval_runtime": 393.8065,
|
| 2039 |
+
"eval_samples_per_second": 161.102,
|
| 2040 |
+
"eval_steps_per_second": 5.035,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 6.683375104427736,
|
| 2045 |
+
"grad_norm": 2.423050880432129,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.8889,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 6.683375104427736,
|
| 2052 |
+
"eval_accuracy": 0.8173146779439671,
|
| 2053 |
+
"eval_loss": 0.9342746138572693,
|
| 2054 |
+
"eval_runtime": 393.0912,
|
| 2055 |
+
"eval_samples_per_second": 161.395,
|
| 2056 |
+
"eval_steps_per_second": 5.045,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 6.735588972431078,
|
| 2061 |
+
"grad_norm": 2.796268939971924,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.8885,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 6.735588972431078,
|
| 2068 |
+
"eval_accuracy": 0.817551689369001,
|
| 2069 |
+
"eval_loss": 0.9292306303977966,
|
| 2070 |
+
"eval_runtime": 394.1087,
|
| 2071 |
+
"eval_samples_per_second": 160.978,
|
| 2072 |
+
"eval_steps_per_second": 5.032,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 6.787802840434419,
|
| 2077 |
+
"grad_norm": 2.8128762245178223,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.8854,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 6.787802840434419,
|
| 2084 |
+
"eval_accuracy": 0.8176226440260885,
|
| 2085 |
+
"eval_loss": 0.9265376925468445,
|
| 2086 |
+
"eval_runtime": 394.2997,
|
| 2087 |
+
"eval_samples_per_second": 160.9,
|
| 2088 |
+
"eval_steps_per_second": 5.029,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 6.840016708437761,
|
| 2093 |
+
"grad_norm": 2.8697590827941895,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.88,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 6.840016708437761,
|
| 2100 |
+
"eval_accuracy": 0.8185052265994757,
|
| 2101 |
+
"eval_loss": 0.928463876247406,
|
| 2102 |
+
"eval_runtime": 394.323,
|
| 2103 |
+
"eval_samples_per_second": 160.891,
|
| 2104 |
+
"eval_steps_per_second": 5.029,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 6.8922305764411025,
|
| 2109 |
+
"grad_norm": 2.6351752281188965,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.8841,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 6.8922305764411025,
|
| 2116 |
+
"eval_accuracy": 0.8183555493481675,
|
| 2117 |
+
"eval_loss": 0.9194802045822144,
|
| 2118 |
+
"eval_runtime": 400.5152,
|
| 2119 |
+
"eval_samples_per_second": 158.403,
|
| 2120 |
+
"eval_steps_per_second": 4.951,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 6.944444444444445,
|
| 2125 |
+
"grad_norm": 2.2477810382843018,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.8807,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 6.944444444444445,
|
| 2132 |
+
"eval_accuracy": 0.8187747090050445,
|
| 2133 |
+
"eval_loss": 0.9138378500938416,
|
| 2134 |
+
"eval_runtime": 394.4731,
|
| 2135 |
+
"eval_samples_per_second": 160.83,
|
| 2136 |
+
"eval_steps_per_second": 5.027,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 6.996658312447786,
|
| 2141 |
+
"grad_norm": 2.1748201847076416,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.8836,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 6.996658312447786,
|
| 2148 |
+
"eval_accuracy": 0.8190003355856446,
|
| 2149 |
+
"eval_loss": 0.9168543815612793,
|
| 2150 |
+
"eval_runtime": 394.6502,
|
| 2151 |
+
"eval_samples_per_second": 160.758,
|
| 2152 |
+
"eval_steps_per_second": 5.025,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 7.048872180451128,
|
| 2157 |
+
"grad_norm": 2.7435665130615234,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.8707,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 7.048872180451128,
|
| 2164 |
+
"eval_accuracy": 0.8191917867246192,
|
| 2165 |
+
"eval_loss": 0.9231117367744446,
|
| 2166 |
+
"eval_runtime": 392.3745,
|
| 2167 |
+
"eval_samples_per_second": 161.69,
|
| 2168 |
+
"eval_steps_per_second": 5.054,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 7.101086048454469,
|
| 2173 |
+
"grad_norm": 2.569931745529175,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.8701,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 7.101086048454469,
|
| 2180 |
+
"eval_accuracy": 0.819969121038317,
|
| 2181 |
+
"eval_loss": 0.9153112173080444,
|
| 2182 |
+
"eval_runtime": 393.0875,
|
| 2183 |
+
"eval_samples_per_second": 161.397,
|
| 2184 |
+
"eval_steps_per_second": 5.045,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 7.1532999164578115,
|
| 2189 |
+
"grad_norm": 2.819972515106201,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.8656,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 7.1532999164578115,
|
| 2196 |
+
"eval_accuracy": 0.8194268014568229,
|
| 2197 |
+
"eval_loss": 0.9146177768707275,
|
| 2198 |
+
"eval_runtime": 394.2167,
|
| 2199 |
+
"eval_samples_per_second": 160.934,
|
| 2200 |
+
"eval_steps_per_second": 5.03,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 7.205513784461153,
|
| 2205 |
+
"grad_norm": 2.415208578109741,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.871,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 7.205513784461153,
|
| 2212 |
+
"eval_accuracy": 0.8198532803574815,
|
| 2213 |
+
"eval_loss": NaN,
|
| 2214 |
+
"eval_runtime": 395.7273,
|
| 2215 |
+
"eval_samples_per_second": 160.32,
|
| 2216 |
+
"eval_steps_per_second": 5.011,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 7.257727652464495,
|
| 2221 |
+
"grad_norm": 2.7900257110595703,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.8685,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 7.257727652464495,
|
| 2228 |
+
"eval_accuracy": 0.8207157152692325,
|
| 2229 |
+
"eval_loss": 0.9127238392829895,
|
| 2230 |
+
"eval_runtime": 392.7333,
|
| 2231 |
+
"eval_samples_per_second": 161.542,
|
| 2232 |
+
"eval_steps_per_second": 5.049,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 7.309941520467836,
|
| 2237 |
+
"grad_norm": 2.643833637237549,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.8623,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 7.309941520467836,
|
| 2244 |
+
"eval_accuracy": 0.8202351749773661,
|
| 2245 |
+
"eval_loss": 0.9171856045722961,
|
| 2246 |
+
"eval_runtime": 393.097,
|
| 2247 |
+
"eval_samples_per_second": 161.393,
|
| 2248 |
+
"eval_steps_per_second": 5.045,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 7.362155388471178,
|
| 2253 |
+
"grad_norm": 2.5037946701049805,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.8601,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 7.362155388471178,
|
| 2260 |
+
"eval_accuracy": 0.8211525544791177,
|
| 2261 |
+
"eval_loss": 0.903904139995575,
|
| 2262 |
+
"eval_runtime": 394.3466,
|
| 2263 |
+
"eval_samples_per_second": 160.881,
|
| 2264 |
+
"eval_steps_per_second": 5.029,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 7.41436925647452,
|
| 2269 |
+
"grad_norm": 2.7093377113342285,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.8573,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 7.41436925647452,
|
| 2276 |
+
"eval_accuracy": 0.8214304681574243,
|
| 2277 |
+
"eval_loss": 0.9036524891853333,
|
| 2278 |
+
"eval_runtime": 393.5778,
|
| 2279 |
+
"eval_samples_per_second": 161.196,
|
| 2280 |
+
"eval_steps_per_second": 5.038,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 7.466583124477861,
|
| 2285 |
+
"grad_norm": 2.760674476623535,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.8633,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 7.466583124477861,
|
| 2292 |
+
"eval_accuracy": 0.8217283854693812,
|
| 2293 |
+
"eval_loss": 0.9000177979469299,
|
| 2294 |
+
"eval_runtime": 396.5628,
|
| 2295 |
+
"eval_samples_per_second": 159.982,
|
| 2296 |
+
"eval_steps_per_second": 5.0,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 7.518796992481203,
|
| 2301 |
+
"grad_norm": 2.831023693084717,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.8576,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 7.518796992481203,
|
| 2308 |
+
"eval_accuracy": 0.8211579127032714,
|
| 2309 |
+
"eval_loss": 0.918953001499176,
|
| 2310 |
+
"eval_runtime": 400.5863,
|
| 2311 |
+
"eval_samples_per_second": 158.375,
|
| 2312 |
+
"eval_steps_per_second": 4.95,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 7.571010860484545,
|
| 2317 |
+
"grad_norm": 2.453390598297119,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.854,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 7.571010860484545,
|
| 2324 |
+
"eval_accuracy": 0.8220418032437955,
|
| 2325 |
+
"eval_loss": 0.9041558504104614,
|
| 2326 |
+
"eval_runtime": 399.249,
|
| 2327 |
+
"eval_samples_per_second": 158.906,
|
| 2328 |
+
"eval_steps_per_second": 4.967,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 7.6232247284878865,
|
| 2333 |
+
"grad_norm": 2.6029062271118164,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.8615,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 7.6232247284878865,
|
| 2340 |
+
"eval_accuracy": 0.8223261586980396,
|
| 2341 |
+
"eval_loss": 0.8994259834289551,
|
| 2342 |
+
"eval_runtime": 387.0421,
|
| 2343 |
+
"eval_samples_per_second": 163.918,
|
| 2344 |
+
"eval_steps_per_second": 5.123,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 7.675438596491228,
|
| 2349 |
+
"grad_norm": 2.3841817378997803,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.8632,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 7.675438596491228,
|
| 2356 |
+
"eval_accuracy": 0.8218723618637837,
|
| 2357 |
+
"eval_loss": 0.9080346822738647,
|
| 2358 |
+
"eval_runtime": 368.6926,
|
| 2359 |
+
"eval_samples_per_second": 172.076,
|
| 2360 |
+
"eval_steps_per_second": 5.378,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 7.72765246449457,
|
| 2365 |
+
"grad_norm": 3.8187203407287598,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.8519,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 7.72765246449457,
|
| 2372 |
+
"eval_accuracy": 0.8223824804514319,
|
| 2373 |
+
"eval_loss": 0.9071189761161804,
|
| 2374 |
+
"eval_runtime": 367.458,
|
| 2375 |
+
"eval_samples_per_second": 172.654,
|
| 2376 |
+
"eval_steps_per_second": 5.397,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 7.779866332497911,
|
| 2381 |
+
"grad_norm": 2.744349241256714,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.8569,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 7.779866332497911,
|
| 2388 |
+
"eval_accuracy": 0.8229439133116854,
|
| 2389 |
+
"eval_loss": 0.9003260731697083,
|
| 2390 |
+
"eval_runtime": 369.188,
|
| 2391 |
+
"eval_samples_per_second": 171.845,
|
| 2392 |
+
"eval_steps_per_second": 5.371,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 7.832080200501253,
|
| 2397 |
+
"grad_norm": 2.3172342777252197,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.8502,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 7.832080200501253,
|
| 2404 |
+
"eval_accuracy": 0.8229715307093431,
|
| 2405 |
+
"eval_loss": 0.9045436382293701,
|
| 2406 |
+
"eval_runtime": 368.1168,
|
| 2407 |
+
"eval_samples_per_second": 172.345,
|
| 2408 |
+
"eval_steps_per_second": 5.387,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 7.884294068504595,
|
| 2413 |
+
"grad_norm": 2.6108663082122803,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.8536,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 7.884294068504595,
|
| 2420 |
+
"eval_accuracy": 0.8234089591247552,
|
| 2421 |
+
"eval_loss": 0.9026762247085571,
|
| 2422 |
+
"eval_runtime": 366.8391,
|
| 2423 |
+
"eval_samples_per_second": 172.945,
|
| 2424 |
+
"eval_steps_per_second": 5.406,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 7.936507936507937,
|
| 2429 |
+
"grad_norm": 2.315197467803955,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.85,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 7.936507936507937,
|
| 2436 |
+
"eval_accuracy": 0.8232359994242249,
|
| 2437 |
+
"eval_loss": 0.8995440006256104,
|
| 2438 |
+
"eval_runtime": 367.8582,
|
| 2439 |
+
"eval_samples_per_second": 172.466,
|
| 2440 |
+
"eval_steps_per_second": 5.391,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 7.988721804511278,
|
| 2445 |
+
"grad_norm": 2.611060380935669,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.8453,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 7.988721804511278,
|
| 2452 |
+
"eval_accuracy": 0.8234605237364002,
|
| 2453 |
+
"eval_loss": 0.8912914395332336,
|
| 2454 |
+
"eval_runtime": 367.6079,
|
| 2455 |
+
"eval_samples_per_second": 172.583,
|
| 2456 |
+
"eval_steps_per_second": 5.394,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 8.04093567251462,
|
| 2461 |
+
"grad_norm": 2.4672865867614746,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.8294,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 8.04093567251462,
|
| 2468 |
+
"eval_accuracy": 0.8244379262655704,
|
| 2469 |
+
"eval_loss": 0.896998405456543,
|
| 2470 |
+
"eval_runtime": 367.8054,
|
| 2471 |
+
"eval_samples_per_second": 172.491,
|
| 2472 |
+
"eval_steps_per_second": 5.391,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 8.093149540517961,
|
| 2477 |
+
"grad_norm": 2.645519495010376,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.849,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 8.093149540517961,
|
| 2484 |
+
"eval_accuracy": 0.8244201499571945,
|
| 2485 |
+
"eval_loss": 0.8957926034927368,
|
| 2486 |
+
"eval_runtime": 369.9508,
|
| 2487 |
+
"eval_samples_per_second": 171.49,
|
| 2488 |
+
"eval_steps_per_second": 5.36,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 8.145363408521304,
|
| 2493 |
+
"grad_norm": 2.7964978218078613,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.8363,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 8.145363408521304,
|
| 2500 |
+
"eval_accuracy": 0.8244232177017738,
|
| 2501 |
+
"eval_loss": 0.8884155750274658,
|
| 2502 |
+
"eval_runtime": 367.358,
|
| 2503 |
+
"eval_samples_per_second": 172.701,
|
| 2504 |
+
"eval_steps_per_second": 5.398,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 8.197577276524646,
|
| 2509 |
+
"grad_norm": 2.48044490814209,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.8355,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 8.197577276524646,
|
| 2516 |
+
"eval_accuracy": 0.82532828384256,
|
| 2517 |
+
"eval_loss": 0.8786827325820923,
|
| 2518 |
+
"eval_runtime": 366.707,
|
| 2519 |
+
"eval_samples_per_second": 173.007,
|
| 2520 |
+
"eval_steps_per_second": 5.408,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 8.249791144527986,
|
| 2525 |
+
"grad_norm": 2.6931161880493164,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.8439,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 8.249791144527986,
|
| 2532 |
+
"eval_accuracy": 0.8246382720646043,
|
| 2533 |
+
"eval_loss": 0.8848527073860168,
|
| 2534 |
+
"eval_runtime": 366.9894,
|
| 2535 |
+
"eval_samples_per_second": 172.874,
|
| 2536 |
+
"eval_steps_per_second": 5.403,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 8.302005012531328,
|
| 2541 |
+
"grad_norm": 2.607593297958374,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.8378,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 8.302005012531328,
|
| 2548 |
+
"eval_accuracy": 0.8252490751457086,
|
| 2549 |
+
"eval_loss": NaN,
|
| 2550 |
+
"eval_runtime": 366.5162,
|
| 2551 |
+
"eval_samples_per_second": 173.097,
|
| 2552 |
+
"eval_steps_per_second": 5.41,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 8.35421888053467,
|
| 2557 |
+
"grad_norm": 2.4364349842071533,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.8322,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 8.35421888053467,
|
| 2564 |
+
"eval_accuracy": 0.8252812537780041,
|
| 2565 |
+
"eval_loss": NaN,
|
| 2566 |
+
"eval_runtime": 367.8653,
|
| 2567 |
+
"eval_samples_per_second": 172.463,
|
| 2568 |
+
"eval_steps_per_second": 5.391,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 8.406432748538013,
|
| 2573 |
+
"grad_norm": 2.541379928588867,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.8366,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 8.406432748538013,
|
| 2580 |
+
"eval_accuracy": 0.8252309925385912,
|
| 2581 |
+
"eval_loss": 0.9011866450309753,
|
| 2582 |
+
"eval_runtime": 367.843,
|
| 2583 |
+
"eval_samples_per_second": 172.473,
|
| 2584 |
+
"eval_steps_per_second": 5.391,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 8.458646616541353,
|
| 2589 |
+
"grad_norm": 2.3451640605926514,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.8313,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 8.458646616541353,
|
| 2596 |
+
"eval_accuracy": 0.8259800217048817,
|
| 2597 |
+
"eval_loss": 0.8837263584136963,
|
| 2598 |
+
"eval_runtime": 367.6187,
|
| 2599 |
+
"eval_samples_per_second": 172.578,
|
| 2600 |
+
"eval_steps_per_second": 5.394,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 8.510860484544695,
|
| 2605 |
+
"grad_norm": 2.6211562156677246,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.8289,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 8.510860484544695,
|
| 2612 |
+
"eval_accuracy": 0.8257132907861338,
|
| 2613 |
+
"eval_loss": 0.8861507773399353,
|
| 2614 |
+
"eval_runtime": 367.3072,
|
| 2615 |
+
"eval_samples_per_second": 172.725,
|
| 2616 |
+
"eval_steps_per_second": 5.399,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 8.563074352548037,
|
| 2621 |
+
"grad_norm": 2.1193125247955322,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.8337,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 8.563074352548037,
|
| 2628 |
+
"eval_accuracy": 0.8263971078151469,
|
| 2629 |
+
"eval_loss": 0.8711854815483093,
|
| 2630 |
+
"eval_runtime": 366.8982,
|
| 2631 |
+
"eval_samples_per_second": 172.917,
|
| 2632 |
+
"eval_steps_per_second": 5.405,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 8.615288220551378,
|
| 2637 |
+
"grad_norm": 2.5682766437530518,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.8207,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 8.615288220551378,
|
| 2644 |
+
"eval_accuracy": 0.8263459457288101,
|
| 2645 |
+
"eval_loss": 0.8814241886138916,
|
| 2646 |
+
"eval_runtime": 366.9346,
|
| 2647 |
+
"eval_samples_per_second": 172.9,
|
| 2648 |
+
"eval_steps_per_second": 5.404,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 8.66750208855472,
|
| 2653 |
+
"grad_norm": 2.460284471511841,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.8273,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 8.66750208855472,
|
| 2660 |
+
"eval_accuracy": 0.8265486834185438,
|
| 2661 |
+
"eval_loss": 0.8726556301116943,
|
| 2662 |
+
"eval_runtime": 367.479,
|
| 2663 |
+
"eval_samples_per_second": 172.644,
|
| 2664 |
+
"eval_steps_per_second": 5.396,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 8.719715956558062,
|
| 2669 |
+
"grad_norm": 2.519878625869751,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.8212,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 8.719715956558062,
|
| 2676 |
+
"eval_accuracy": 0.8265830712467154,
|
| 2677 |
+
"eval_loss": 0.8803038001060486,
|
| 2678 |
+
"eval_runtime": 368.0421,
|
| 2679 |
+
"eval_samples_per_second": 172.38,
|
| 2680 |
+
"eval_steps_per_second": 5.388,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 8.771929824561404,
|
| 2685 |
+
"grad_norm": 2.9779090881347656,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.8267,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 8.771929824561404,
|
| 2692 |
+
"eval_accuracy": 0.8266352426703358,
|
| 2693 |
+
"eval_loss": NaN,
|
| 2694 |
+
"eval_runtime": 370.1684,
|
| 2695 |
+
"eval_samples_per_second": 171.39,
|
| 2696 |
+
"eval_steps_per_second": 5.357,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 8.824143692564745,
|
| 2701 |
+
"grad_norm": 2.5584726333618164,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.8236,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 8.824143692564745,
|
| 2708 |
+
"eval_accuracy": 0.8272979799699072,
|
| 2709 |
+
"eval_loss": 0.8709605932235718,
|
| 2710 |
+
"eval_runtime": 367.2687,
|
| 2711 |
+
"eval_samples_per_second": 172.743,
|
| 2712 |
+
"eval_steps_per_second": 5.399,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 8.876357560568087,
|
| 2717 |
+
"grad_norm": 2.8752248287200928,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.8284,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 8.876357560568087,
|
| 2724 |
+
"eval_accuracy": 0.8270593176674245,
|
| 2725 |
+
"eval_loss": 0.8703105449676514,
|
| 2726 |
+
"eval_runtime": 367.121,
|
| 2727 |
+
"eval_samples_per_second": 172.812,
|
| 2728 |
+
"eval_steps_per_second": 5.401,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 8.928571428571429,
|
| 2733 |
+
"grad_norm": 2.441423177719116,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.8305,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 8.928571428571429,
|
| 2740 |
+
"eval_accuracy": 0.8273055974821587,
|
| 2741 |
+
"eval_loss": 0.8782740831375122,
|
| 2742 |
+
"eval_runtime": 367.2074,
|
| 2743 |
+
"eval_samples_per_second": 172.772,
|
| 2744 |
+
"eval_steps_per_second": 5.4,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 8.980785296574771,
|
| 2749 |
+
"grad_norm": 2.5627357959747314,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.824,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 8.980785296574771,
|
| 2756 |
+
"eval_accuracy": 0.8278787560948533,
|
| 2757 |
+
"eval_loss": 0.8767423033714294,
|
| 2758 |
+
"eval_runtime": 367.8162,
|
| 2759 |
+
"eval_samples_per_second": 172.486,
|
| 2760 |
+
"eval_steps_per_second": 5.391,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 9.032999164578111,
|
| 2765 |
+
"grad_norm": 2.9659039974212646,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.8111,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 9.032999164578111,
|
| 2772 |
+
"eval_accuracy": 0.8281429418603345,
|
| 2773 |
+
"eval_loss": NaN,
|
| 2774 |
+
"eval_runtime": 365.8491,
|
| 2775 |
+
"eval_samples_per_second": 173.413,
|
| 2776 |
+
"eval_steps_per_second": 5.42,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 9.085213032581454,
|
| 2781 |
+
"grad_norm": 2.6677966117858887,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.8114,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 9.085213032581454,
|
| 2788 |
+
"eval_accuracy": 0.8280896401382127,
|
| 2789 |
+
"eval_loss": 0.8696718215942383,
|
| 2790 |
+
"eval_runtime": 368.939,
|
| 2791 |
+
"eval_samples_per_second": 171.961,
|
| 2792 |
+
"eval_steps_per_second": 5.375,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 9.137426900584796,
|
| 2797 |
+
"grad_norm": 2.8159141540527344,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.8214,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 9.137426900584796,
|
| 2804 |
+
"eval_accuracy": 0.8287721660611849,
|
| 2805 |
+
"eval_loss": 0.865568995475769,
|
| 2806 |
+
"eval_runtime": 367.5282,
|
| 2807 |
+
"eval_samples_per_second": 172.621,
|
| 2808 |
+
"eval_steps_per_second": 5.396,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 9.189640768588138,
|
| 2813 |
+
"grad_norm": 2.44439959526062,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.8117,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 9.189640768588138,
|
| 2820 |
+
"eval_accuracy": 0.8286521507089889,
|
| 2821 |
+
"eval_loss": 0.8658538460731506,
|
| 2822 |
+
"eval_runtime": 365.9254,
|
| 2823 |
+
"eval_samples_per_second": 173.377,
|
| 2824 |
+
"eval_steps_per_second": 5.419,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 9.241854636591478,
|
| 2829 |
+
"grad_norm": 2.686903953552246,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.8176,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 9.241854636591478,
|
| 2836 |
+
"eval_accuracy": 0.8286974413686677,
|
| 2837 |
+
"eval_loss": 0.8724528551101685,
|
| 2838 |
+
"eval_runtime": 366.9175,
|
| 2839 |
+
"eval_samples_per_second": 172.908,
|
| 2840 |
+
"eval_steps_per_second": 5.404,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 9.29406850459482,
|
| 2845 |
+
"grad_norm": 2.9780402183532715,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.8098,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 9.29406850459482,
|
| 2852 |
+
"eval_accuracy": 0.8288113105821936,
|
| 2853 |
+
"eval_loss": 0.8693479895591736,
|
| 2854 |
+
"eval_runtime": 366.5706,
|
| 2855 |
+
"eval_samples_per_second": 173.072,
|
| 2856 |
+
"eval_steps_per_second": 5.41,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 9.346282372598163,
|
| 2861 |
+
"grad_norm": 2.3616678714752197,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.8101,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 9.346282372598163,
|
| 2868 |
+
"eval_accuracy": 0.8288534544077049,
|
| 2869 |
+
"eval_loss": 0.8676702380180359,
|
| 2870 |
+
"eval_runtime": 367.6956,
|
| 2871 |
+
"eval_samples_per_second": 172.542,
|
| 2872 |
+
"eval_steps_per_second": 5.393,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 9.398496240601503,
|
| 2877 |
+
"grad_norm": 2.4607014656066895,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.8074,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 9.398496240601503,
|
| 2884 |
+
"eval_accuracy": 0.8293368416792034,
|
| 2885 |
+
"eval_loss": 0.868602454662323,
|
| 2886 |
+
"eval_runtime": 370.3428,
|
| 2887 |
+
"eval_samples_per_second": 171.309,
|
| 2888 |
+
"eval_steps_per_second": 5.354,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 9.450710108604845,
|
| 2893 |
+
"grad_norm": 2.303865432739258,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.8075,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 9.450710108604845,
|
| 2900 |
+
"eval_accuracy": 0.829557819292624,
|
| 2901 |
+
"eval_loss": 0.8760720491409302,
|
| 2902 |
+
"eval_runtime": 366.9515,
|
| 2903 |
+
"eval_samples_per_second": 172.892,
|
| 2904 |
+
"eval_steps_per_second": 5.404,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 9.502923976608187,
|
| 2909 |
+
"grad_norm": 3.1800107955932617,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.813,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 9.502923976608187,
|
| 2916 |
+
"eval_accuracy": 0.8293719470428488,
|
| 2917 |
+
"eval_loss": 0.8725053668022156,
|
| 2918 |
+
"eval_runtime": 366.7268,
|
| 2919 |
+
"eval_samples_per_second": 172.998,
|
| 2920 |
+
"eval_steps_per_second": 5.407,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 9.55513784461153,
|
| 2925 |
+
"grad_norm": 2.895458698272705,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.8142,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 9.55513784461153,
|
| 2932 |
+
"eval_accuracy": 0.8299493767535929,
|
| 2933 |
+
"eval_loss": 0.8597660660743713,
|
| 2934 |
+
"eval_runtime": 366.9591,
|
| 2935 |
+
"eval_samples_per_second": 172.888,
|
| 2936 |
+
"eval_steps_per_second": 5.404,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 9.60735171261487,
|
| 2941 |
+
"grad_norm": 2.494898796081543,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.8028,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 9.60735171261487,
|
| 2948 |
+
"eval_accuracy": 0.8296925250842329,
|
| 2949 |
+
"eval_loss": 0.8663123250007629,
|
| 2950 |
+
"eval_runtime": 366.9071,
|
| 2951 |
+
"eval_samples_per_second": 172.913,
|
| 2952 |
+
"eval_steps_per_second": 5.405,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 9.659565580618212,
|
| 2957 |
+
"grad_norm": 2.6649532318115234,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.8051,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 9.659565580618212,
|
| 2964 |
+
"eval_accuracy": 0.8297183234153672,
|
| 2965 |
+
"eval_loss": 0.8609718084335327,
|
| 2966 |
+
"eval_runtime": 369.3381,
|
| 2967 |
+
"eval_samples_per_second": 171.775,
|
| 2968 |
+
"eval_steps_per_second": 5.369,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 9.711779448621554,
|
| 2973 |
+
"grad_norm": 2.708948850631714,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.7982,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 9.711779448621554,
|
| 2980 |
+
"eval_accuracy": 0.8295886536989987,
|
| 2981 |
+
"eval_loss": 0.8697434067726135,
|
| 2982 |
+
"eval_runtime": 367.9632,
|
| 2983 |
+
"eval_samples_per_second": 172.417,
|
| 2984 |
+
"eval_steps_per_second": 5.389,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 9.763993316624896,
|
| 2989 |
+
"grad_norm": 2.320215940475464,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.7985,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 9.763993316624896,
|
| 2996 |
+
"eval_accuracy": 0.8301079266760504,
|
| 2997 |
+
"eval_loss": 0.8655940890312195,
|
| 2998 |
+
"eval_runtime": 366.2895,
|
| 2999 |
+
"eval_samples_per_second": 173.205,
|
| 3000 |
+
"eval_steps_per_second": 5.414,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 9.816207184628237,
|
| 3005 |
+
"grad_norm": 2.776233673095703,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.8016,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 9.816207184628237,
|
| 3012 |
+
"eval_accuracy": 0.8302324108540118,
|
| 3013 |
+
"eval_loss": 0.8625151515007019,
|
| 3014 |
+
"eval_runtime": 367.8541,
|
| 3015 |
+
"eval_samples_per_second": 172.468,
|
| 3016 |
+
"eval_steps_per_second": 5.391,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 9.868421052631579,
|
| 3021 |
+
"grad_norm": 2.461749792098999,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.8044,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 9.868421052631579,
|
| 3028 |
+
"eval_accuracy": 0.8303814560823992,
|
| 3029 |
+
"eval_loss": NaN,
|
| 3030 |
+
"eval_runtime": 366.9598,
|
| 3031 |
+
"eval_samples_per_second": 172.888,
|
| 3032 |
+
"eval_steps_per_second": 5.404,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 9.920634920634921,
|
| 3037 |
+
"grad_norm": 2.611931085586548,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.8,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 9.920634920634921,
|
| 3044 |
+
"eval_accuracy": 0.8306366115947627,
|
| 3045 |
+
"eval_loss": 0.8623033165931702,
|
| 3046 |
+
"eval_runtime": 368.1962,
|
| 3047 |
+
"eval_samples_per_second": 172.308,
|
| 3048 |
+
"eval_steps_per_second": 5.386,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 9.972848788638263,
|
| 3053 |
+
"grad_norm": 2.500162363052368,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.7972,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 9.972848788638263,
|
| 3060 |
+
"eval_accuracy": 0.8306005889087491,
|
| 3061 |
+
"eval_loss": 0.8600591421127319,
|
| 3062 |
+
"eval_runtime": 369.0733,
|
| 3063 |
+
"eval_samples_per_second": 171.898,
|
| 3064 |
+
"eval_steps_per_second": 5.373,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 10.025062656641603,
|
| 3069 |
+
"grad_norm": 3.495044231414795,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.7929,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 10.025062656641603,
|
| 3076 |
+
"eval_accuracy": 0.8308683319034997,
|
| 3077 |
+
"eval_loss": 0.8547109961509705,
|
| 3078 |
+
"eval_runtime": 368.4222,
|
| 3079 |
+
"eval_samples_per_second": 172.202,
|
| 3080 |
+
"eval_steps_per_second": 5.382,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 10.077276524644946,
|
| 3085 |
+
"grad_norm": 2.7070469856262207,
|
| 3086 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 3087 |
+
"loss": 0.7903,
|
| 3088 |
+
"step": 96500
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 10.077276524644946,
|
| 3092 |
+
"eval_accuracy": 0.8307715292279126,
|
| 3093 |
+
"eval_loss": 0.8600655198097229,
|
| 3094 |
+
"eval_runtime": 366.9772,
|
| 3095 |
+
"eval_samples_per_second": 172.88,
|
| 3096 |
+
"eval_steps_per_second": 5.404,
|
| 3097 |
+
"step": 96500
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 10.129490392648288,
|
| 3101 |
+
"grad_norm": 3.4624171257019043,
|
| 3102 |
+
"learning_rate": 3e-06,
|
| 3103 |
+
"loss": 0.7898,
|
| 3104 |
+
"step": 97000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 10.129490392648288,
|
| 3108 |
+
"eval_accuracy": 0.831032096060647,
|
| 3109 |
+
"eval_loss": 0.853934109210968,
|
| 3110 |
+
"eval_runtime": 368.4043,
|
| 3111 |
+
"eval_samples_per_second": 172.21,
|
| 3112 |
+
"eval_steps_per_second": 5.383,
|
| 3113 |
+
"step": 97000
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 10.18170426065163,
|
| 3117 |
+
"grad_norm": 2.6027026176452637,
|
| 3118 |
+
"learning_rate": 2.5e-06,
|
| 3119 |
+
"loss": 0.7994,
|
| 3120 |
+
"step": 97500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 10.18170426065163,
|
| 3124 |
+
"eval_accuracy": 0.8308584971832438,
|
| 3125 |
+
"eval_loss": 0.8530024886131287,
|
| 3126 |
+
"eval_runtime": 366.5565,
|
| 3127 |
+
"eval_samples_per_second": 173.078,
|
| 3128 |
+
"eval_steps_per_second": 5.41,
|
| 3129 |
+
"step": 97500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 10.23391812865497,
|
| 3133 |
+
"grad_norm": 2.5377743244171143,
|
| 3134 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 3135 |
+
"loss": 0.7926,
|
| 3136 |
+
"step": 98000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 10.23391812865497,
|
| 3140 |
+
"eval_accuracy": 0.8310957564926591,
|
| 3141 |
+
"eval_loss": 0.8588021993637085,
|
| 3142 |
+
"eval_runtime": 366.0793,
|
| 3143 |
+
"eval_samples_per_second": 173.304,
|
| 3144 |
+
"eval_steps_per_second": 5.417,
|
| 3145 |
+
"step": 98000
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 10.286131996658312,
|
| 3149 |
+
"grad_norm": 2.9179139137268066,
|
| 3150 |
+
"learning_rate": 1.5e-06,
|
| 3151 |
+
"loss": 0.785,
|
| 3152 |
+
"step": 98500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 10.286131996658312,
|
| 3156 |
+
"eval_accuracy": 0.8314594660073593,
|
| 3157 |
+
"eval_loss": 0.8528650403022766,
|
| 3158 |
+
"eval_runtime": 366.8402,
|
| 3159 |
+
"eval_samples_per_second": 172.945,
|
| 3160 |
+
"eval_steps_per_second": 5.406,
|
| 3161 |
+
"step": 98500
|
| 3162 |
+
}
|
| 3163 |
+
],
|
| 3164 |
+
"logging_steps": 500,
|
| 3165 |
+
"max_steps": 100000,
|
| 3166 |
+
"num_input_tokens_seen": 0,
|
| 3167 |
+
"num_train_epochs": 11,
|
| 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": 8.304812876669911e+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 sha256:71160b812b1ac68a1cc3d833569a3f1f519d777db07a56cf7f682d12dff9f302
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|