Uploading checkpoint-99000 for mbert - npi-deva
Browse files- README.md +67 -0
- config.json +37 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +3201 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: google-bert/bert-base-multilingual-cased
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: npi-Deva
|
| 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 |
+
# npi-Deva
|
| 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: 0.3886
|
| 22 |
+
- Accuracy: 0.9047
|
| 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:a4f454f9c73dcd41735e7e234607df85224a33d48a89f285082e04c00c52736d
|
| 3 |
+
size 711921812
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee0e6efe7350ebd6cc4f26ede311547643b6baed0c7af14930c2fb406181be9e
|
| 3 |
+
size 1423965114
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba5c000ab9dec45a95d45a6e133b107bb2260833e14c8a9b7d47b365738c625e
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ea567b1093528972c430b63a5bf5d6f5b265876f9bc0a19b392383be83c77d1
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4415404b182ca8f419ce636856a8efa57ef9cce9de682e39e40579cf7bbd5cfa
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": false,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3867943286895752,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/mbert/npi-Deva/checkpoint-99000",
|
| 4 |
+
"epoch": 22.307345651194233,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 99000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11266336187471834,
|
| 13 |
+
"grad_norm": 2.965751886367798,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.2301,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11266336187471834,
|
| 20 |
+
"eval_accuracy": 0.773184921040326,
|
| 21 |
+
"eval_loss": 1.0424206256866455,
|
| 22 |
+
"eval_runtime": 220.277,
|
| 23 |
+
"eval_samples_per_second": 125.542,
|
| 24 |
+
"eval_steps_per_second": 3.927,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.22532672374943669,
|
| 29 |
+
"grad_norm": 2.740410327911377,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.0543,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.22532672374943669,
|
| 36 |
+
"eval_accuracy": 0.7909380298838988,
|
| 37 |
+
"eval_loss": 0.9471855163574219,
|
| 38 |
+
"eval_runtime": 220.5287,
|
| 39 |
+
"eval_samples_per_second": 125.399,
|
| 40 |
+
"eval_steps_per_second": 3.922,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.337990085624155,
|
| 45 |
+
"grad_norm": 2.7863168716430664,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 0.9779,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.337990085624155,
|
| 52 |
+
"eval_accuracy": 0.8025380313108481,
|
| 53 |
+
"eval_loss": 0.8843335509300232,
|
| 54 |
+
"eval_runtime": 220.1694,
|
| 55 |
+
"eval_samples_per_second": 125.603,
|
| 56 |
+
"eval_steps_per_second": 3.929,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.45065344749887337,
|
| 61 |
+
"grad_norm": 2.5414557456970215,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 0.926,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.45065344749887337,
|
| 68 |
+
"eval_accuracy": 0.8101656739468518,
|
| 69 |
+
"eval_loss": 0.8543083667755127,
|
| 70 |
+
"eval_runtime": 220.7797,
|
| 71 |
+
"eval_samples_per_second": 125.256,
|
| 72 |
+
"eval_steps_per_second": 3.918,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.5633168093735917,
|
| 77 |
+
"grad_norm": 2.377737283706665,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 0.8865,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.5633168093735917,
|
| 84 |
+
"eval_accuracy": 0.8173040702951884,
|
| 85 |
+
"eval_loss": 0.810703456401825,
|
| 86 |
+
"eval_runtime": 220.8366,
|
| 87 |
+
"eval_samples_per_second": 125.224,
|
| 88 |
+
"eval_steps_per_second": 3.917,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.67598017124831,
|
| 93 |
+
"grad_norm": 2.3488988876342773,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 0.8609,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.67598017124831,
|
| 100 |
+
"eval_accuracy": 0.821973743458109,
|
| 101 |
+
"eval_loss": 0.7787520289421082,
|
| 102 |
+
"eval_runtime": 221.2787,
|
| 103 |
+
"eval_samples_per_second": 124.974,
|
| 104 |
+
"eval_steps_per_second": 3.909,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.7886435331230284,
|
| 109 |
+
"grad_norm": 2.1220295429229736,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 0.8342,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.7886435331230284,
|
| 116 |
+
"eval_accuracy": 0.8262216641689282,
|
| 117 |
+
"eval_loss": 0.762144923210144,
|
| 118 |
+
"eval_runtime": 220.8289,
|
| 119 |
+
"eval_samples_per_second": 125.228,
|
| 120 |
+
"eval_steps_per_second": 3.917,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9013068949977467,
|
| 125 |
+
"grad_norm": 2.0968008041381836,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 0.819,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.9013068949977467,
|
| 132 |
+
"eval_accuracy": 0.8298798163116309,
|
| 133 |
+
"eval_loss": 0.7466955184936523,
|
| 134 |
+
"eval_runtime": 220.013,
|
| 135 |
+
"eval_samples_per_second": 125.693,
|
| 136 |
+
"eval_steps_per_second": 3.932,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.0139702568724651,
|
| 141 |
+
"grad_norm": 2.1498773097991943,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 0.7979,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.0139702568724651,
|
| 148 |
+
"eval_accuracy": 0.8318434647099575,
|
| 149 |
+
"eval_loss": 0.7348815202713013,
|
| 150 |
+
"eval_runtime": 220.8229,
|
| 151 |
+
"eval_samples_per_second": 125.232,
|
| 152 |
+
"eval_steps_per_second": 3.917,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.1266336187471835,
|
| 157 |
+
"grad_norm": 2.163381576538086,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 0.7814,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.1266336187471835,
|
| 164 |
+
"eval_accuracy": 0.8349864178467281,
|
| 165 |
+
"eval_loss": 0.7180664539337158,
|
| 166 |
+
"eval_runtime": 219.7442,
|
| 167 |
+
"eval_samples_per_second": 125.846,
|
| 168 |
+
"eval_steps_per_second": 3.936,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.2392969806219019,
|
| 173 |
+
"grad_norm": 2.431119680404663,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 0.7665,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.2392969806219019,
|
| 180 |
+
"eval_accuracy": 0.8369151584278837,
|
| 181 |
+
"eval_loss": 0.7159287333488464,
|
| 182 |
+
"eval_runtime": 220.9495,
|
| 183 |
+
"eval_samples_per_second": 125.16,
|
| 184 |
+
"eval_steps_per_second": 3.915,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.35196034249662,
|
| 189 |
+
"grad_norm": 2.2182135581970215,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 0.7555,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.35196034249662,
|
| 196 |
+
"eval_accuracy": 0.8390386817390197,
|
| 197 |
+
"eval_loss": 0.6999027132987976,
|
| 198 |
+
"eval_runtime": 221.2257,
|
| 199 |
+
"eval_samples_per_second": 125.004,
|
| 200 |
+
"eval_steps_per_second": 3.91,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.4646237043713384,
|
| 205 |
+
"grad_norm": 2.1569323539733887,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 0.7479,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.4646237043713384,
|
| 212 |
+
"eval_accuracy": 0.8422681179825285,
|
| 213 |
+
"eval_loss": 0.689199686050415,
|
| 214 |
+
"eval_runtime": 222.1646,
|
| 215 |
+
"eval_samples_per_second": 124.475,
|
| 216 |
+
"eval_steps_per_second": 3.894,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.5772870662460567,
|
| 221 |
+
"grad_norm": 2.1323976516723633,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.733,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.5772870662460567,
|
| 228 |
+
"eval_accuracy": 0.8439326355101854,
|
| 229 |
+
"eval_loss": 0.6770957112312317,
|
| 230 |
+
"eval_runtime": 220.0047,
|
| 231 |
+
"eval_samples_per_second": 125.697,
|
| 232 |
+
"eval_steps_per_second": 3.932,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.6899504281207751,
|
| 237 |
+
"grad_norm": 5.134857177734375,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.7254,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.6899504281207751,
|
| 244 |
+
"eval_accuracy": 0.8453398865939418,
|
| 245 |
+
"eval_loss": 0.6668263077735901,
|
| 246 |
+
"eval_runtime": 220.3226,
|
| 247 |
+
"eval_samples_per_second": 125.516,
|
| 248 |
+
"eval_steps_per_second": 3.926,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.8026137899954935,
|
| 253 |
+
"grad_norm": 2.0616443157196045,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.7155,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.8026137899954935,
|
| 260 |
+
"eval_accuracy": 0.8463415002486427,
|
| 261 |
+
"eval_loss": 0.6613638997077942,
|
| 262 |
+
"eval_runtime": 221.4149,
|
| 263 |
+
"eval_samples_per_second": 124.897,
|
| 264 |
+
"eval_steps_per_second": 3.907,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.9152771518702119,
|
| 269 |
+
"grad_norm": 1.8310041427612305,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.7057,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.9152771518702119,
|
| 276 |
+
"eval_accuracy": 0.8482940895875467,
|
| 277 |
+
"eval_loss": 0.6528915762901306,
|
| 278 |
+
"eval_runtime": 222.056,
|
| 279 |
+
"eval_samples_per_second": 124.536,
|
| 280 |
+
"eval_steps_per_second": 3.895,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.0279405137449302,
|
| 285 |
+
"grad_norm": 1.93686842918396,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.7005,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.0279405137449302,
|
| 292 |
+
"eval_accuracy": 0.849507663539711,
|
| 293 |
+
"eval_loss": 0.6522949934005737,
|
| 294 |
+
"eval_runtime": 221.8006,
|
| 295 |
+
"eval_samples_per_second": 124.68,
|
| 296 |
+
"eval_steps_per_second": 3.9,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.1406038756196484,
|
| 301 |
+
"grad_norm": 6.06415319442749,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.6884,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.1406038756196484,
|
| 308 |
+
"eval_accuracy": 0.8502759566677828,
|
| 309 |
+
"eval_loss": 0.6491975784301758,
|
| 310 |
+
"eval_runtime": 221.9876,
|
| 311 |
+
"eval_samples_per_second": 124.575,
|
| 312 |
+
"eval_steps_per_second": 3.897,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.253267237494367,
|
| 317 |
+
"grad_norm": 1.9235719442367554,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.6821,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.253267237494367,
|
| 324 |
+
"eval_accuracy": 0.8524909295282275,
|
| 325 |
+
"eval_loss": 0.6343050599098206,
|
| 326 |
+
"eval_runtime": 222.2817,
|
| 327 |
+
"eval_samples_per_second": 124.41,
|
| 328 |
+
"eval_steps_per_second": 3.891,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.365930599369085,
|
| 333 |
+
"grad_norm": 1.8421759605407715,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.6767,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.365930599369085,
|
| 340 |
+
"eval_accuracy": 0.8534436334949597,
|
| 341 |
+
"eval_loss": 0.623904287815094,
|
| 342 |
+
"eval_runtime": 221.9784,
|
| 343 |
+
"eval_samples_per_second": 124.58,
|
| 344 |
+
"eval_steps_per_second": 3.897,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.4785939612438037,
|
| 349 |
+
"grad_norm": 1.9507330656051636,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.6792,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.4785939612438037,
|
| 356 |
+
"eval_accuracy": 0.8552298873542783,
|
| 357 |
+
"eval_loss": 0.6220438480377197,
|
| 358 |
+
"eval_runtime": 221.6861,
|
| 359 |
+
"eval_samples_per_second": 124.744,
|
| 360 |
+
"eval_steps_per_second": 3.902,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.591257323118522,
|
| 365 |
+
"grad_norm": 2.11086106300354,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.668,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.591257323118522,
|
| 372 |
+
"eval_accuracy": 0.8557371543230742,
|
| 373 |
+
"eval_loss": 0.6222216486930847,
|
| 374 |
+
"eval_runtime": 221.2659,
|
| 375 |
+
"eval_samples_per_second": 124.981,
|
| 376 |
+
"eval_steps_per_second": 3.909,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.70392068499324,
|
| 381 |
+
"grad_norm": 2.1847715377807617,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.6636,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.70392068499324,
|
| 388 |
+
"eval_accuracy": 0.8559949812795903,
|
| 389 |
+
"eval_loss": 0.6197636127471924,
|
| 390 |
+
"eval_runtime": 222.3204,
|
| 391 |
+
"eval_samples_per_second": 124.388,
|
| 392 |
+
"eval_steps_per_second": 3.891,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.8165840468679586,
|
| 397 |
+
"grad_norm": 2.1351499557495117,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.6576,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.8165840468679586,
|
| 404 |
+
"eval_accuracy": 0.8577423671742627,
|
| 405 |
+
"eval_loss": 0.6103814840316772,
|
| 406 |
+
"eval_runtime": 222.1201,
|
| 407 |
+
"eval_samples_per_second": 124.5,
|
| 408 |
+
"eval_steps_per_second": 3.894,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.9292474087426768,
|
| 413 |
+
"grad_norm": 3.9510111808776855,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.6488,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.9292474087426768,
|
| 420 |
+
"eval_accuracy": 0.858561338833408,
|
| 421 |
+
"eval_loss": 0.6049174070358276,
|
| 422 |
+
"eval_runtime": 221.9998,
|
| 423 |
+
"eval_samples_per_second": 124.568,
|
| 424 |
+
"eval_steps_per_second": 3.896,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 3.0419107706173953,
|
| 429 |
+
"grad_norm": 1.8234397172927856,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.6438,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 3.0419107706173953,
|
| 436 |
+
"eval_accuracy": 0.8591052959636053,
|
| 437 |
+
"eval_loss": 0.6051846742630005,
|
| 438 |
+
"eval_runtime": 220.9713,
|
| 439 |
+
"eval_samples_per_second": 125.147,
|
| 440 |
+
"eval_steps_per_second": 3.915,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 3.1545741324921135,
|
| 445 |
+
"grad_norm": 1.9275134801864624,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.6369,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 3.1545741324921135,
|
| 452 |
+
"eval_accuracy": 0.8599297280864646,
|
| 453 |
+
"eval_loss": 0.6021236181259155,
|
| 454 |
+
"eval_runtime": 222.1682,
|
| 455 |
+
"eval_samples_per_second": 124.473,
|
| 456 |
+
"eval_steps_per_second": 3.893,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.267237494366832,
|
| 461 |
+
"grad_norm": 2.4342575073242188,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.6375,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.267237494366832,
|
| 468 |
+
"eval_accuracy": 0.8612232782302242,
|
| 469 |
+
"eval_loss": 0.5935059785842896,
|
| 470 |
+
"eval_runtime": 221.4028,
|
| 471 |
+
"eval_samples_per_second": 124.904,
|
| 472 |
+
"eval_steps_per_second": 3.907,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.3799008562415502,
|
| 477 |
+
"grad_norm": 1.8208547830581665,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.6327,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 3.3799008562415502,
|
| 484 |
+
"eval_accuracy": 0.8619705169680766,
|
| 485 |
+
"eval_loss": 0.5865727663040161,
|
| 486 |
+
"eval_runtime": 221.4519,
|
| 487 |
+
"eval_samples_per_second": 124.876,
|
| 488 |
+
"eval_steps_per_second": 3.906,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.492564218116269,
|
| 493 |
+
"grad_norm": 1.8497122526168823,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.6289,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 3.492564218116269,
|
| 500 |
+
"eval_accuracy": 0.8624703434485368,
|
| 501 |
+
"eval_loss": 0.5854940414428711,
|
| 502 |
+
"eval_runtime": 222.074,
|
| 503 |
+
"eval_samples_per_second": 124.526,
|
| 504 |
+
"eval_steps_per_second": 3.895,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 3.605227579990987,
|
| 509 |
+
"grad_norm": 1.7389825582504272,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.6231,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.605227579990987,
|
| 516 |
+
"eval_accuracy": 0.8635307164001665,
|
| 517 |
+
"eval_loss": 0.5809486508369446,
|
| 518 |
+
"eval_runtime": 222.3436,
|
| 519 |
+
"eval_samples_per_second": 124.375,
|
| 520 |
+
"eval_steps_per_second": 3.89,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 3.717890941865705,
|
| 525 |
+
"grad_norm": 1.7109190225601196,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.6193,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 3.717890941865705,
|
| 532 |
+
"eval_accuracy": 0.8642588913962003,
|
| 533 |
+
"eval_loss": 0.5757493376731873,
|
| 534 |
+
"eval_runtime": 220.862,
|
| 535 |
+
"eval_samples_per_second": 125.209,
|
| 536 |
+
"eval_steps_per_second": 3.916,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.8305543037404237,
|
| 541 |
+
"grad_norm": 2.09114408493042,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.619,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 3.8305543037404237,
|
| 548 |
+
"eval_accuracy": 0.8644031427528578,
|
| 549 |
+
"eval_loss": 0.5797725319862366,
|
| 550 |
+
"eval_runtime": 220.9835,
|
| 551 |
+
"eval_samples_per_second": 125.141,
|
| 552 |
+
"eval_steps_per_second": 3.914,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 3.943217665615142,
|
| 557 |
+
"grad_norm": 6.745112419128418,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.6127,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 3.943217665615142,
|
| 564 |
+
"eval_accuracy": 0.8645245282957764,
|
| 565 |
+
"eval_loss": 0.5759025812149048,
|
| 566 |
+
"eval_runtime": 222.2291,
|
| 567 |
+
"eval_samples_per_second": 124.439,
|
| 568 |
+
"eval_steps_per_second": 3.892,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 4.0558810274898605,
|
| 573 |
+
"grad_norm": 1.7710591554641724,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.6081,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 4.0558810274898605,
|
| 580 |
+
"eval_accuracy": 0.8658915432042757,
|
| 581 |
+
"eval_loss": 0.5714759230613708,
|
| 582 |
+
"eval_runtime": 221.6135,
|
| 583 |
+
"eval_samples_per_second": 124.785,
|
| 584 |
+
"eval_steps_per_second": 3.903,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 4.168544389364579,
|
| 589 |
+
"grad_norm": 1.8267593383789062,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.5988,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 4.168544389364579,
|
| 596 |
+
"eval_accuracy": 0.8665697779685045,
|
| 597 |
+
"eval_loss": 0.5671255588531494,
|
| 598 |
+
"eval_runtime": 221.0373,
|
| 599 |
+
"eval_samples_per_second": 125.11,
|
| 600 |
+
"eval_steps_per_second": 3.913,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 4.281207751239297,
|
| 605 |
+
"grad_norm": 1.6686463356018066,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.5981,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 4.281207751239297,
|
| 612 |
+
"eval_accuracy": 0.8667210799508446,
|
| 613 |
+
"eval_loss": 0.5654014348983765,
|
| 614 |
+
"eval_runtime": 221.1716,
|
| 615 |
+
"eval_samples_per_second": 125.034,
|
| 616 |
+
"eval_steps_per_second": 3.911,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 4.393871113114015,
|
| 621 |
+
"grad_norm": 1.6965349912643433,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.599,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 4.393871113114015,
|
| 628 |
+
"eval_accuracy": 0.8677269725072129,
|
| 629 |
+
"eval_loss": 0.5655470490455627,
|
| 630 |
+
"eval_runtime": 221.4343,
|
| 631 |
+
"eval_samples_per_second": 124.886,
|
| 632 |
+
"eval_steps_per_second": 3.906,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 4.506534474988734,
|
| 637 |
+
"grad_norm": 1.653952956199646,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.5976,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 4.506534474988734,
|
| 644 |
+
"eval_accuracy": 0.8685987876288259,
|
| 645 |
+
"eval_loss": 0.5560412406921387,
|
| 646 |
+
"eval_runtime": 220.5715,
|
| 647 |
+
"eval_samples_per_second": 125.374,
|
| 648 |
+
"eval_steps_per_second": 3.922,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 4.619197836863452,
|
| 653 |
+
"grad_norm": 1.7568910121917725,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.5941,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 4.619197836863452,
|
| 660 |
+
"eval_accuracy": 0.868412802308659,
|
| 661 |
+
"eval_loss": 0.5624808669090271,
|
| 662 |
+
"eval_runtime": 220.7945,
|
| 663 |
+
"eval_samples_per_second": 125.248,
|
| 664 |
+
"eval_steps_per_second": 3.918,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 4.73186119873817,
|
| 669 |
+
"grad_norm": 1.7545663118362427,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.5871,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 4.73186119873817,
|
| 676 |
+
"eval_accuracy": 0.8700149406874658,
|
| 677 |
+
"eval_loss": 0.5546574592590332,
|
| 678 |
+
"eval_runtime": 220.2428,
|
| 679 |
+
"eval_samples_per_second": 125.561,
|
| 680 |
+
"eval_steps_per_second": 3.927,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.844524560612888,
|
| 685 |
+
"grad_norm": 1.9459997415542603,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.5891,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 4.844524560612888,
|
| 692 |
+
"eval_accuracy": 0.8703311716376315,
|
| 693 |
+
"eval_loss": 0.5456222295761108,
|
| 694 |
+
"eval_runtime": 220.5867,
|
| 695 |
+
"eval_samples_per_second": 125.366,
|
| 696 |
+
"eval_steps_per_second": 3.921,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 4.957187922487607,
|
| 701 |
+
"grad_norm": 1.9034132957458496,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.5828,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 4.957187922487607,
|
| 708 |
+
"eval_accuracy": 0.8704027728365514,
|
| 709 |
+
"eval_loss": 0.549776554107666,
|
| 710 |
+
"eval_runtime": 221.4908,
|
| 711 |
+
"eval_samples_per_second": 124.854,
|
| 712 |
+
"eval_steps_per_second": 3.905,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 5.069851284362326,
|
| 717 |
+
"grad_norm": 1.881596565246582,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.5767,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 5.069851284362326,
|
| 724 |
+
"eval_accuracy": 0.8711589106147363,
|
| 725 |
+
"eval_loss": 0.5461272597312927,
|
| 726 |
+
"eval_runtime": 220.457,
|
| 727 |
+
"eval_samples_per_second": 125.439,
|
| 728 |
+
"eval_steps_per_second": 3.924,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 5.182514646237044,
|
| 733 |
+
"grad_norm": 1.9157260656356812,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.5731,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 5.182514646237044,
|
| 740 |
+
"eval_accuracy": 0.871975417070376,
|
| 741 |
+
"eval_loss": 0.5400785207748413,
|
| 742 |
+
"eval_runtime": 220.8692,
|
| 743 |
+
"eval_samples_per_second": 125.205,
|
| 744 |
+
"eval_steps_per_second": 3.916,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 5.295178008111762,
|
| 749 |
+
"grad_norm": 1.9823201894760132,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.5736,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 5.295178008111762,
|
| 756 |
+
"eval_accuracy": 0.8723751389743424,
|
| 757 |
+
"eval_loss": 0.5401638746261597,
|
| 758 |
+
"eval_runtime": 221.6042,
|
| 759 |
+
"eval_samples_per_second": 124.79,
|
| 760 |
+
"eval_steps_per_second": 3.903,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 5.40784136998648,
|
| 765 |
+
"grad_norm": 1.905613660812378,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.5747,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 5.40784136998648,
|
| 772 |
+
"eval_accuracy": 0.8724923660478054,
|
| 773 |
+
"eval_loss": 0.5441656112670898,
|
| 774 |
+
"eval_runtime": 221.3067,
|
| 775 |
+
"eval_samples_per_second": 124.958,
|
| 776 |
+
"eval_steps_per_second": 3.909,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 5.520504731861199,
|
| 781 |
+
"grad_norm": 1.5278126001358032,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.5681,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 5.520504731861199,
|
| 788 |
+
"eval_accuracy": 0.8728878650306285,
|
| 789 |
+
"eval_loss": 0.538100004196167,
|
| 790 |
+
"eval_runtime": 222.0369,
|
| 791 |
+
"eval_samples_per_second": 124.547,
|
| 792 |
+
"eval_steps_per_second": 3.896,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 5.633168093735917,
|
| 797 |
+
"grad_norm": 1.6478660106658936,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.5658,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 5.633168093735917,
|
| 804 |
+
"eval_accuracy": 0.8736624848239579,
|
| 805 |
+
"eval_loss": 0.5357881784439087,
|
| 806 |
+
"eval_runtime": 220.4147,
|
| 807 |
+
"eval_samples_per_second": 125.463,
|
| 808 |
+
"eval_steps_per_second": 3.924,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 5.745831455610635,
|
| 813 |
+
"grad_norm": 3.0473523139953613,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.5644,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 5.745831455610635,
|
| 820 |
+
"eval_accuracy": 0.8743903767129565,
|
| 821 |
+
"eval_loss": 0.5344362854957581,
|
| 822 |
+
"eval_runtime": 221.5481,
|
| 823 |
+
"eval_samples_per_second": 124.822,
|
| 824 |
+
"eval_steps_per_second": 3.904,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.8584948174853535,
|
| 829 |
+
"grad_norm": 1.8053028583526611,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.5622,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 5.8584948174853535,
|
| 836 |
+
"eval_accuracy": 0.874178054098396,
|
| 837 |
+
"eval_loss": 0.5315510630607605,
|
| 838 |
+
"eval_runtime": 221.4537,
|
| 839 |
+
"eval_samples_per_second": 124.875,
|
| 840 |
+
"eval_steps_per_second": 3.906,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 5.9711581793600725,
|
| 845 |
+
"grad_norm": 1.5863131284713745,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.5578,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 5.9711581793600725,
|
| 852 |
+
"eval_accuracy": 0.8753070050808498,
|
| 853 |
+
"eval_loss": 0.5271232724189758,
|
| 854 |
+
"eval_runtime": 221.5103,
|
| 855 |
+
"eval_samples_per_second": 124.843,
|
| 856 |
+
"eval_steps_per_second": 3.905,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 6.083821541234791,
|
| 861 |
+
"grad_norm": 1.7924689054489136,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.5546,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 6.083821541234791,
|
| 868 |
+
"eval_accuracy": 0.8749559789605048,
|
| 869 |
+
"eval_loss": 0.5305372476577759,
|
| 870 |
+
"eval_runtime": 220.6828,
|
| 871 |
+
"eval_samples_per_second": 125.311,
|
| 872 |
+
"eval_steps_per_second": 3.92,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 6.196484903109509,
|
| 877 |
+
"grad_norm": 1.6176671981811523,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.5553,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 6.196484903109509,
|
| 884 |
+
"eval_accuracy": 0.8752024294778373,
|
| 885 |
+
"eval_loss": 0.5255776047706604,
|
| 886 |
+
"eval_runtime": 220.8919,
|
| 887 |
+
"eval_samples_per_second": 125.192,
|
| 888 |
+
"eval_steps_per_second": 3.916,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 6.309148264984227,
|
| 893 |
+
"grad_norm": 1.855047583580017,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.5506,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 6.309148264984227,
|
| 900 |
+
"eval_accuracy": 0.8761331460452507,
|
| 901 |
+
"eval_loss": 0.52358478307724,
|
| 902 |
+
"eval_runtime": 220.7028,
|
| 903 |
+
"eval_samples_per_second": 125.3,
|
| 904 |
+
"eval_steps_per_second": 3.919,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 6.421811626858945,
|
| 909 |
+
"grad_norm": 1.6553348302841187,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.5439,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 6.421811626858945,
|
| 916 |
+
"eval_accuracy": 0.8768096662621753,
|
| 917 |
+
"eval_loss": 0.5175614953041077,
|
| 918 |
+
"eval_runtime": 221.3868,
|
| 919 |
+
"eval_samples_per_second": 124.913,
|
| 920 |
+
"eval_steps_per_second": 3.907,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 6.534474988733664,
|
| 925 |
+
"grad_norm": 1.8099743127822876,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.5486,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 6.534474988733664,
|
| 932 |
+
"eval_accuracy": 0.8767345488093528,
|
| 933 |
+
"eval_loss": 0.5191013216972351,
|
| 934 |
+
"eval_runtime": 221.0646,
|
| 935 |
+
"eval_samples_per_second": 125.095,
|
| 936 |
+
"eval_steps_per_second": 3.913,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 6.647138350608382,
|
| 941 |
+
"grad_norm": 1.7723827362060547,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.5442,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 6.647138350608382,
|
| 948 |
+
"eval_accuracy": 0.8777180592418201,
|
| 949 |
+
"eval_loss": 0.5211535096168518,
|
| 950 |
+
"eval_runtime": 222.034,
|
| 951 |
+
"eval_samples_per_second": 124.548,
|
| 952 |
+
"eval_steps_per_second": 3.896,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 6.7598017124831005,
|
| 957 |
+
"grad_norm": 1.7134077548980713,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.5412,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 6.7598017124831005,
|
| 964 |
+
"eval_accuracy": 0.8771853054768167,
|
| 965 |
+
"eval_loss": 0.5161250829696655,
|
| 966 |
+
"eval_runtime": 221.7362,
|
| 967 |
+
"eval_samples_per_second": 124.716,
|
| 968 |
+
"eval_steps_per_second": 3.901,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.872465074357819,
|
| 973 |
+
"grad_norm": 1.7683045864105225,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.5402,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 6.872465074357819,
|
| 980 |
+
"eval_accuracy": 0.8773839402820733,
|
| 981 |
+
"eval_loss": 0.5139411687850952,
|
| 982 |
+
"eval_runtime": 220.8209,
|
| 983 |
+
"eval_samples_per_second": 125.233,
|
| 984 |
+
"eval_steps_per_second": 3.917,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 6.985128436232538,
|
| 989 |
+
"grad_norm": 1.8624660968780518,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.5395,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 6.985128436232538,
|
| 996 |
+
"eval_accuracy": 0.8783624777319803,
|
| 997 |
+
"eval_loss": 0.5147821307182312,
|
| 998 |
+
"eval_runtime": 221.2924,
|
| 999 |
+
"eval_samples_per_second": 124.966,
|
| 1000 |
+
"eval_steps_per_second": 3.909,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 7.097791798107256,
|
| 1005 |
+
"grad_norm": 1.6134588718414307,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.5323,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 7.097791798107256,
|
| 1012 |
+
"eval_accuracy": 0.8783227617479554,
|
| 1013 |
+
"eval_loss": 0.5111725330352783,
|
| 1014 |
+
"eval_runtime": 221.33,
|
| 1015 |
+
"eval_samples_per_second": 124.945,
|
| 1016 |
+
"eval_steps_per_second": 3.908,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 7.210455159981974,
|
| 1021 |
+
"grad_norm": 1.8190521001815796,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.5341,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 7.210455159981974,
|
| 1028 |
+
"eval_accuracy": 0.8789992011172492,
|
| 1029 |
+
"eval_loss": 0.5084385275840759,
|
| 1030 |
+
"eval_runtime": 221.6067,
|
| 1031 |
+
"eval_samples_per_second": 124.789,
|
| 1032 |
+
"eval_steps_per_second": 3.903,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 7.323118521856692,
|
| 1037 |
+
"grad_norm": 1.7497199773788452,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.5325,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 7.323118521856692,
|
| 1044 |
+
"eval_accuracy": 0.8800775404890228,
|
| 1045 |
+
"eval_loss": 0.5039363503456116,
|
| 1046 |
+
"eval_runtime": 222.1191,
|
| 1047 |
+
"eval_samples_per_second": 124.501,
|
| 1048 |
+
"eval_steps_per_second": 3.894,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 7.43578188373141,
|
| 1053 |
+
"grad_norm": 1.6325268745422363,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.5309,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 7.43578188373141,
|
| 1060 |
+
"eval_accuracy": 0.8801297464169966,
|
| 1061 |
+
"eval_loss": 0.505262017250061,
|
| 1062 |
+
"eval_runtime": 221.5433,
|
| 1063 |
+
"eval_samples_per_second": 124.824,
|
| 1064 |
+
"eval_steps_per_second": 3.904,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 7.548445245606128,
|
| 1069 |
+
"grad_norm": 1.7531828880310059,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.5283,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 7.548445245606128,
|
| 1076 |
+
"eval_accuracy": 0.8802065110814512,
|
| 1077 |
+
"eval_loss": 0.5030723810195923,
|
| 1078 |
+
"eval_runtime": 221.3827,
|
| 1079 |
+
"eval_samples_per_second": 124.915,
|
| 1080 |
+
"eval_steps_per_second": 3.907,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 7.661108607480847,
|
| 1085 |
+
"grad_norm": 1.7174723148345947,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.5254,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 7.661108607480847,
|
| 1092 |
+
"eval_accuracy": 0.8808179417817528,
|
| 1093 |
+
"eval_loss": 0.5008535385131836,
|
| 1094 |
+
"eval_runtime": 220.9595,
|
| 1095 |
+
"eval_samples_per_second": 125.154,
|
| 1096 |
+
"eval_steps_per_second": 3.915,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 7.773771969355566,
|
| 1101 |
+
"grad_norm": 1.4874796867370605,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.5217,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 7.773771969355566,
|
| 1108 |
+
"eval_accuracy": 0.8805426353661477,
|
| 1109 |
+
"eval_loss": 0.5023674368858337,
|
| 1110 |
+
"eval_runtime": 221.8005,
|
| 1111 |
+
"eval_samples_per_second": 124.68,
|
| 1112 |
+
"eval_steps_per_second": 3.9,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 7.886435331230284,
|
| 1117 |
+
"grad_norm": 1.721684455871582,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.5268,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 7.886435331230284,
|
| 1124 |
+
"eval_accuracy": 0.8809166125769631,
|
| 1125 |
+
"eval_loss": 0.49913424253463745,
|
| 1126 |
+
"eval_runtime": 221.9778,
|
| 1127 |
+
"eval_samples_per_second": 124.58,
|
| 1128 |
+
"eval_steps_per_second": 3.897,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 7.999098693105002,
|
| 1133 |
+
"grad_norm": 1.6311215162277222,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.5279,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 7.999098693105002,
|
| 1140 |
+
"eval_accuracy": 0.8812218745311298,
|
| 1141 |
+
"eval_loss": 0.49702906608581543,
|
| 1142 |
+
"eval_runtime": 221.8919,
|
| 1143 |
+
"eval_samples_per_second": 124.628,
|
| 1144 |
+
"eval_steps_per_second": 3.898,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 8.111762054979721,
|
| 1149 |
+
"grad_norm": 1.843680739402771,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.5113,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 8.111762054979721,
|
| 1156 |
+
"eval_accuracy": 0.8819420345077154,
|
| 1157 |
+
"eval_loss": 0.4924590289592743,
|
| 1158 |
+
"eval_runtime": 220.9686,
|
| 1159 |
+
"eval_samples_per_second": 125.149,
|
| 1160 |
+
"eval_steps_per_second": 3.915,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 8.224425416854439,
|
| 1165 |
+
"grad_norm": 2.186274290084839,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.519,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 8.224425416854439,
|
| 1172 |
+
"eval_accuracy": 0.8821893135016634,
|
| 1173 |
+
"eval_loss": 0.49259641766548157,
|
| 1174 |
+
"eval_runtime": 221.6028,
|
| 1175 |
+
"eval_samples_per_second": 124.791,
|
| 1176 |
+
"eval_steps_per_second": 3.903,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 8.337088778729157,
|
| 1181 |
+
"grad_norm": 1.6842992305755615,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.5142,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 8.337088778729157,
|
| 1188 |
+
"eval_accuracy": 0.8831766874688345,
|
| 1189 |
+
"eval_loss": 0.49090540409088135,
|
| 1190 |
+
"eval_runtime": 221.5262,
|
| 1191 |
+
"eval_samples_per_second": 124.834,
|
| 1192 |
+
"eval_steps_per_second": 3.905,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 8.449752140603875,
|
| 1197 |
+
"grad_norm": 1.69620943069458,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.5118,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 8.449752140603875,
|
| 1204 |
+
"eval_accuracy": 0.8829115546267233,
|
| 1205 |
+
"eval_loss": 0.48741188645362854,
|
| 1206 |
+
"eval_runtime": 220.7417,
|
| 1207 |
+
"eval_samples_per_second": 125.278,
|
| 1208 |
+
"eval_steps_per_second": 3.919,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 8.562415502478594,
|
| 1213 |
+
"grad_norm": 1.7428566217422485,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.513,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 8.562415502478594,
|
| 1220 |
+
"eval_accuracy": 0.883224147742304,
|
| 1221 |
+
"eval_loss": 0.4888308644294739,
|
| 1222 |
+
"eval_runtime": 221.978,
|
| 1223 |
+
"eval_samples_per_second": 124.58,
|
| 1224 |
+
"eval_steps_per_second": 3.897,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 8.675078864353312,
|
| 1229 |
+
"grad_norm": 1.7894220352172852,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.5158,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 8.675078864353312,
|
| 1236 |
+
"eval_accuracy": 0.8831726096386725,
|
| 1237 |
+
"eval_loss": 0.48892539739608765,
|
| 1238 |
+
"eval_runtime": 221.6541,
|
| 1239 |
+
"eval_samples_per_second": 124.762,
|
| 1240 |
+
"eval_steps_per_second": 3.902,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 8.78774222622803,
|
| 1245 |
+
"grad_norm": 1.8389184474945068,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.5074,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 8.78774222622803,
|
| 1252 |
+
"eval_accuracy": 0.8836604617832933,
|
| 1253 |
+
"eval_loss": 0.4904680550098419,
|
| 1254 |
+
"eval_runtime": 220.5534,
|
| 1255 |
+
"eval_samples_per_second": 125.385,
|
| 1256 |
+
"eval_steps_per_second": 3.922,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 8.90040558810275,
|
| 1261 |
+
"grad_norm": 1.6285669803619385,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.5115,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 8.90040558810275,
|
| 1268 |
+
"eval_accuracy": 0.8834682744822249,
|
| 1269 |
+
"eval_loss": 0.4873930513858795,
|
| 1270 |
+
"eval_runtime": 221.8348,
|
| 1271 |
+
"eval_samples_per_second": 124.66,
|
| 1272 |
+
"eval_steps_per_second": 3.899,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 9.013068949977468,
|
| 1277 |
+
"grad_norm": 1.892903447151184,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.506,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 9.013068949977468,
|
| 1284 |
+
"eval_accuracy": 0.8844121521679462,
|
| 1285 |
+
"eval_loss": 0.48553282022476196,
|
| 1286 |
+
"eval_runtime": 221.0439,
|
| 1287 |
+
"eval_samples_per_second": 125.106,
|
| 1288 |
+
"eval_steps_per_second": 3.913,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 9.125732311852186,
|
| 1293 |
+
"grad_norm": 1.5161460638046265,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.5006,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 9.125732311852186,
|
| 1300 |
+
"eval_accuracy": 0.8845845321702142,
|
| 1301 |
+
"eval_loss": 0.4854166805744171,
|
| 1302 |
+
"eval_runtime": 221.5376,
|
| 1303 |
+
"eval_samples_per_second": 124.828,
|
| 1304 |
+
"eval_steps_per_second": 3.905,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 9.238395673726904,
|
| 1309 |
+
"grad_norm": 1.6559338569641113,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.4998,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 9.238395673726904,
|
| 1316 |
+
"eval_accuracy": 0.8846773258713508,
|
| 1317 |
+
"eval_loss": 0.47937873005867004,
|
| 1318 |
+
"eval_runtime": 221.7432,
|
| 1319 |
+
"eval_samples_per_second": 124.712,
|
| 1320 |
+
"eval_steps_per_second": 3.901,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 9.351059035601622,
|
| 1325 |
+
"grad_norm": 1.5425843000411987,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.4993,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 9.351059035601622,
|
| 1332 |
+
"eval_accuracy": 0.8852505184740784,
|
| 1333 |
+
"eval_loss": 0.47994357347488403,
|
| 1334 |
+
"eval_runtime": 220.8861,
|
| 1335 |
+
"eval_samples_per_second": 125.196,
|
| 1336 |
+
"eval_steps_per_second": 3.916,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 9.46372239747634,
|
| 1341 |
+
"grad_norm": 1.6957345008850098,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.4978,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 9.46372239747634,
|
| 1348 |
+
"eval_accuracy": 0.8847172732012654,
|
| 1349 |
+
"eval_loss": 0.48131656646728516,
|
| 1350 |
+
"eval_runtime": 222.0591,
|
| 1351 |
+
"eval_samples_per_second": 124.534,
|
| 1352 |
+
"eval_steps_per_second": 3.895,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 9.576385759351059,
|
| 1357 |
+
"grad_norm": 1.9139741659164429,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.4989,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 9.576385759351059,
|
| 1364 |
+
"eval_accuracy": 0.8862352978048973,
|
| 1365 |
+
"eval_loss": 0.4748667776584625,
|
| 1366 |
+
"eval_runtime": 221.9766,
|
| 1367 |
+
"eval_samples_per_second": 124.581,
|
| 1368 |
+
"eval_steps_per_second": 3.897,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 9.689049121225777,
|
| 1373 |
+
"grad_norm": 1.770585536956787,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.4974,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 9.689049121225777,
|
| 1380 |
+
"eval_accuracy": 0.8855722252421147,
|
| 1381 |
+
"eval_loss": 0.4763648211956024,
|
| 1382 |
+
"eval_runtime": 220.554,
|
| 1383 |
+
"eval_samples_per_second": 125.384,
|
| 1384 |
+
"eval_steps_per_second": 3.922,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 9.801712483100495,
|
| 1389 |
+
"grad_norm": 1.6551371812820435,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.4978,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 9.801712483100495,
|
| 1396 |
+
"eval_accuracy": 0.8858765050235756,
|
| 1397 |
+
"eval_loss": 0.47770920395851135,
|
| 1398 |
+
"eval_runtime": 221.8932,
|
| 1399 |
+
"eval_samples_per_second": 124.628,
|
| 1400 |
+
"eval_steps_per_second": 3.898,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 9.914375844975215,
|
| 1405 |
+
"grad_norm": 1.6118969917297363,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.4942,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 9.914375844975215,
|
| 1412 |
+
"eval_accuracy": 0.8865845660569847,
|
| 1413 |
+
"eval_loss": 0.47676002979278564,
|
| 1414 |
+
"eval_runtime": 221.7004,
|
| 1415 |
+
"eval_samples_per_second": 124.736,
|
| 1416 |
+
"eval_steps_per_second": 3.902,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 10.027039206849933,
|
| 1421 |
+
"grad_norm": 1.8588035106658936,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.4955,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 10.027039206849933,
|
| 1428 |
+
"eval_accuracy": 0.8870998796760368,
|
| 1429 |
+
"eval_loss": 0.47594934701919556,
|
| 1430 |
+
"eval_runtime": 221.976,
|
| 1431 |
+
"eval_samples_per_second": 124.581,
|
| 1432 |
+
"eval_steps_per_second": 3.897,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 10.139702568724651,
|
| 1437 |
+
"grad_norm": 1.6966643333435059,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.489,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 10.139702568724651,
|
| 1444 |
+
"eval_accuracy": 0.8869448016018396,
|
| 1445 |
+
"eval_loss": 0.477344274520874,
|
| 1446 |
+
"eval_runtime": 221.5008,
|
| 1447 |
+
"eval_samples_per_second": 124.848,
|
| 1448 |
+
"eval_steps_per_second": 3.905,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 10.25236593059937,
|
| 1453 |
+
"grad_norm": 1.7615017890930176,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.4849,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 10.25236593059937,
|
| 1460 |
+
"eval_accuracy": 0.8868306586288885,
|
| 1461 |
+
"eval_loss": 0.4725435972213745,
|
| 1462 |
+
"eval_runtime": 221.7608,
|
| 1463 |
+
"eval_samples_per_second": 124.702,
|
| 1464 |
+
"eval_steps_per_second": 3.901,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 10.365029292474087,
|
| 1469 |
+
"grad_norm": 1.7889434099197388,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.4818,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 10.365029292474087,
|
| 1476 |
+
"eval_accuracy": 0.887833900017014,
|
| 1477 |
+
"eval_loss": 0.4671822190284729,
|
| 1478 |
+
"eval_runtime": 222.1274,
|
| 1479 |
+
"eval_samples_per_second": 124.496,
|
| 1480 |
+
"eval_steps_per_second": 3.894,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 10.477692654348806,
|
| 1485 |
+
"grad_norm": 1.7761868238449097,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.4864,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 10.477692654348806,
|
| 1492 |
+
"eval_accuracy": 0.887966177980069,
|
| 1493 |
+
"eval_loss": 0.46516725420951843,
|
| 1494 |
+
"eval_runtime": 221.4768,
|
| 1495 |
+
"eval_samples_per_second": 124.862,
|
| 1496 |
+
"eval_steps_per_second": 3.906,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 10.590356016223524,
|
| 1501 |
+
"grad_norm": 1.7193918228149414,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.4854,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 10.590356016223524,
|
| 1508 |
+
"eval_accuracy": 0.8878875431862944,
|
| 1509 |
+
"eval_loss": 0.4649243652820587,
|
| 1510 |
+
"eval_runtime": 221.9203,
|
| 1511 |
+
"eval_samples_per_second": 124.612,
|
| 1512 |
+
"eval_steps_per_second": 3.898,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 10.703019378098242,
|
| 1517 |
+
"grad_norm": 1.681303858757019,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.4842,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 10.703019378098242,
|
| 1524 |
+
"eval_accuracy": 0.8880860212733241,
|
| 1525 |
+
"eval_loss": 0.4627833366394043,
|
| 1526 |
+
"eval_runtime": 220.7325,
|
| 1527 |
+
"eval_samples_per_second": 125.283,
|
| 1528 |
+
"eval_steps_per_second": 3.919,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 10.81568273997296,
|
| 1533 |
+
"grad_norm": 1.689483642578125,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.4853,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 10.81568273997296,
|
| 1540 |
+
"eval_accuracy": 0.8884850427627177,
|
| 1541 |
+
"eval_loss": 0.4670482873916626,
|
| 1542 |
+
"eval_runtime": 222.2087,
|
| 1543 |
+
"eval_samples_per_second": 124.451,
|
| 1544 |
+
"eval_steps_per_second": 3.893,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 10.92834610184768,
|
| 1549 |
+
"grad_norm": 1.6489872932434082,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.4825,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 10.92834610184768,
|
| 1556 |
+
"eval_accuracy": 0.8886944679602043,
|
| 1557 |
+
"eval_loss": 0.4673362970352173,
|
| 1558 |
+
"eval_runtime": 221.8366,
|
| 1559 |
+
"eval_samples_per_second": 124.659,
|
| 1560 |
+
"eval_steps_per_second": 3.899,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 11.041009463722398,
|
| 1565 |
+
"grad_norm": 1.6207237243652344,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.4783,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 11.041009463722398,
|
| 1572 |
+
"eval_accuracy": 0.8887642317859056,
|
| 1573 |
+
"eval_loss": 0.46382275223731995,
|
| 1574 |
+
"eval_runtime": 222.0203,
|
| 1575 |
+
"eval_samples_per_second": 124.556,
|
| 1576 |
+
"eval_steps_per_second": 3.896,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 11.153672825597116,
|
| 1581 |
+
"grad_norm": 1.7849069833755493,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.4755,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 11.153672825597116,
|
| 1588 |
+
"eval_accuracy": 0.8889988225245398,
|
| 1589 |
+
"eval_loss": 0.4611697793006897,
|
| 1590 |
+
"eval_runtime": 221.8086,
|
| 1591 |
+
"eval_samples_per_second": 124.675,
|
| 1592 |
+
"eval_steps_per_second": 3.9,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 11.266336187471834,
|
| 1597 |
+
"grad_norm": 1.7341585159301758,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.4766,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 11.266336187471834,
|
| 1604 |
+
"eval_accuracy": 0.8896719975387671,
|
| 1605 |
+
"eval_loss": 0.45947107672691345,
|
| 1606 |
+
"eval_runtime": 221.6153,
|
| 1607 |
+
"eval_samples_per_second": 124.784,
|
| 1608 |
+
"eval_steps_per_second": 3.903,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 11.378999549346553,
|
| 1613 |
+
"grad_norm": 1.6157374382019043,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.4758,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 11.378999549346553,
|
| 1620 |
+
"eval_accuracy": 0.8899909483321304,
|
| 1621 |
+
"eval_loss": 0.4591013193130493,
|
| 1622 |
+
"eval_runtime": 221.8952,
|
| 1623 |
+
"eval_samples_per_second": 124.626,
|
| 1624 |
+
"eval_steps_per_second": 3.898,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 11.49166291122127,
|
| 1629 |
+
"grad_norm": 1.3931312561035156,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.4749,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 11.49166291122127,
|
| 1636 |
+
"eval_accuracy": 0.8898331334878133,
|
| 1637 |
+
"eval_loss": 0.4599143862724304,
|
| 1638 |
+
"eval_runtime": 221.5141,
|
| 1639 |
+
"eval_samples_per_second": 124.841,
|
| 1640 |
+
"eval_steps_per_second": 3.905,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 11.604326273095989,
|
| 1645 |
+
"grad_norm": 1.5027562379837036,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.4696,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 11.604326273095989,
|
| 1652 |
+
"eval_accuracy": 0.8903252192404275,
|
| 1653 |
+
"eval_loss": 0.4557996988296509,
|
| 1654 |
+
"eval_runtime": 221.4742,
|
| 1655 |
+
"eval_samples_per_second": 124.863,
|
| 1656 |
+
"eval_steps_per_second": 3.906,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 11.716989634970707,
|
| 1661 |
+
"grad_norm": 2.007624864578247,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.4731,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 11.716989634970707,
|
| 1668 |
+
"eval_accuracy": 0.8906047731898101,
|
| 1669 |
+
"eval_loss": 0.4601598381996155,
|
| 1670 |
+
"eval_runtime": 221.6161,
|
| 1671 |
+
"eval_samples_per_second": 124.783,
|
| 1672 |
+
"eval_steps_per_second": 3.903,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 11.829652996845425,
|
| 1677 |
+
"grad_norm": 1.623124361038208,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.4705,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 11.829652996845425,
|
| 1684 |
+
"eval_accuracy": 0.8907063641623542,
|
| 1685 |
+
"eval_loss": 0.4568343460559845,
|
| 1686 |
+
"eval_runtime": 221.7063,
|
| 1687 |
+
"eval_samples_per_second": 124.733,
|
| 1688 |
+
"eval_steps_per_second": 3.902,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 11.942316358720145,
|
| 1693 |
+
"grad_norm": 1.7550790309906006,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.4712,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 11.942316358720145,
|
| 1700 |
+
"eval_accuracy": 0.8906701808811146,
|
| 1701 |
+
"eval_loss": 0.4544416666030884,
|
| 1702 |
+
"eval_runtime": 221.7786,
|
| 1703 |
+
"eval_samples_per_second": 124.692,
|
| 1704 |
+
"eval_steps_per_second": 3.9,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 12.054979720594863,
|
| 1709 |
+
"grad_norm": 1.8783979415893555,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.4672,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 12.054979720594863,
|
| 1716 |
+
"eval_accuracy": 0.8910758036453728,
|
| 1717 |
+
"eval_loss": 0.45520085096359253,
|
| 1718 |
+
"eval_runtime": 221.6441,
|
| 1719 |
+
"eval_samples_per_second": 124.768,
|
| 1720 |
+
"eval_steps_per_second": 3.903,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 12.167643082469581,
|
| 1725 |
+
"grad_norm": 1.7316193580627441,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.4643,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 12.167643082469581,
|
| 1732 |
+
"eval_accuracy": 0.8910305824271204,
|
| 1733 |
+
"eval_loss": 0.4555051028728485,
|
| 1734 |
+
"eval_runtime": 221.45,
|
| 1735 |
+
"eval_samples_per_second": 124.877,
|
| 1736 |
+
"eval_steps_per_second": 3.906,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 12.2803064443443,
|
| 1741 |
+
"grad_norm": 1.6475858688354492,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.4634,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 12.2803064443443,
|
| 1748 |
+
"eval_accuracy": 0.8916132904164534,
|
| 1749 |
+
"eval_loss": 0.450579971075058,
|
| 1750 |
+
"eval_runtime": 221.3289,
|
| 1751 |
+
"eval_samples_per_second": 124.945,
|
| 1752 |
+
"eval_steps_per_second": 3.908,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 12.392969806219018,
|
| 1757 |
+
"grad_norm": 1.6666234731674194,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.4629,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 12.392969806219018,
|
| 1764 |
+
"eval_accuracy": 0.8920182501631405,
|
| 1765 |
+
"eval_loss": 0.4492991268634796,
|
| 1766 |
+
"eval_runtime": 220.8234,
|
| 1767 |
+
"eval_samples_per_second": 125.231,
|
| 1768 |
+
"eval_steps_per_second": 3.917,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 12.505633168093736,
|
| 1773 |
+
"grad_norm": 2.040255308151245,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.4577,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 12.505633168093736,
|
| 1780 |
+
"eval_accuracy": 0.8917890008025184,
|
| 1781 |
+
"eval_loss": 0.45352259278297424,
|
| 1782 |
+
"eval_runtime": 222.1131,
|
| 1783 |
+
"eval_samples_per_second": 124.504,
|
| 1784 |
+
"eval_steps_per_second": 3.894,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 12.618296529968454,
|
| 1789 |
+
"grad_norm": 1.6200906038284302,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.4597,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 12.618296529968454,
|
| 1796 |
+
"eval_accuracy": 0.8922171868098442,
|
| 1797 |
+
"eval_loss": 0.45517975091934204,
|
| 1798 |
+
"eval_runtime": 221.5067,
|
| 1799 |
+
"eval_samples_per_second": 124.845,
|
| 1800 |
+
"eval_steps_per_second": 3.905,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 12.730959891843172,
|
| 1805 |
+
"grad_norm": 1.8632248640060425,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.4624,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 12.730959891843172,
|
| 1812 |
+
"eval_accuracy": 0.8927531961352251,
|
| 1813 |
+
"eval_loss": 0.44637200236320496,
|
| 1814 |
+
"eval_runtime": 221.488,
|
| 1815 |
+
"eval_samples_per_second": 124.856,
|
| 1816 |
+
"eval_steps_per_second": 3.905,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 12.84362325371789,
|
| 1821 |
+
"grad_norm": 1.6908427476882935,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.46,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 12.84362325371789,
|
| 1828 |
+
"eval_accuracy": 0.8920658168036726,
|
| 1829 |
+
"eval_loss": 0.44909459352493286,
|
| 1830 |
+
"eval_runtime": 221.6242,
|
| 1831 |
+
"eval_samples_per_second": 124.779,
|
| 1832 |
+
"eval_steps_per_second": 3.903,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 12.95628661559261,
|
| 1837 |
+
"grad_norm": 1.7786799669265747,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.4586,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 12.95628661559261,
|
| 1844 |
+
"eval_accuracy": 0.8929494005257145,
|
| 1845 |
+
"eval_loss": 0.4447159469127655,
|
| 1846 |
+
"eval_runtime": 221.7737,
|
| 1847 |
+
"eval_samples_per_second": 124.695,
|
| 1848 |
+
"eval_steps_per_second": 3.9,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 13.068949977467328,
|
| 1853 |
+
"grad_norm": 1.7628467082977295,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.4558,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 13.068949977467328,
|
| 1860 |
+
"eval_accuracy": 0.8926064267317521,
|
| 1861 |
+
"eval_loss": 0.4458833336830139,
|
| 1862 |
+
"eval_runtime": 221.4227,
|
| 1863 |
+
"eval_samples_per_second": 124.892,
|
| 1864 |
+
"eval_steps_per_second": 3.907,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 13.181613339342046,
|
| 1869 |
+
"grad_norm": 1.5658234357833862,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.4542,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 13.181613339342046,
|
| 1876 |
+
"eval_accuracy": 0.8932430818063928,
|
| 1877 |
+
"eval_loss": 0.4461354613304138,
|
| 1878 |
+
"eval_runtime": 221.5061,
|
| 1879 |
+
"eval_samples_per_second": 124.845,
|
| 1880 |
+
"eval_steps_per_second": 3.905,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 13.294276701216765,
|
| 1885 |
+
"grad_norm": 1.5327554941177368,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.455,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 13.294276701216765,
|
| 1892 |
+
"eval_accuracy": 0.8931742171282963,
|
| 1893 |
+
"eval_loss": 0.4385415017604828,
|
| 1894 |
+
"eval_runtime": 220.6336,
|
| 1895 |
+
"eval_samples_per_second": 125.339,
|
| 1896 |
+
"eval_steps_per_second": 3.921,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 13.406940063091483,
|
| 1901 |
+
"grad_norm": 1.804396390914917,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.4506,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 13.406940063091483,
|
| 1908 |
+
"eval_accuracy": 0.8937421063264991,
|
| 1909 |
+
"eval_loss": 0.4429979622364044,
|
| 1910 |
+
"eval_runtime": 221.8018,
|
| 1911 |
+
"eval_samples_per_second": 124.679,
|
| 1912 |
+
"eval_steps_per_second": 3.9,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 13.519603424966201,
|
| 1917 |
+
"grad_norm": 1.8558369874954224,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.4542,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 13.519603424966201,
|
| 1924 |
+
"eval_accuracy": 0.8935920533223497,
|
| 1925 |
+
"eval_loss": 0.4469524025917053,
|
| 1926 |
+
"eval_runtime": 220.934,
|
| 1927 |
+
"eval_samples_per_second": 125.169,
|
| 1928 |
+
"eval_steps_per_second": 3.915,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 13.632266786840919,
|
| 1933 |
+
"grad_norm": 1.7201515436172485,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.4535,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 13.632266786840919,
|
| 1940 |
+
"eval_accuracy": 0.8939343684906524,
|
| 1941 |
+
"eval_loss": 0.44035276770591736,
|
| 1942 |
+
"eval_runtime": 220.7553,
|
| 1943 |
+
"eval_samples_per_second": 125.27,
|
| 1944 |
+
"eval_steps_per_second": 3.918,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 13.744930148715637,
|
| 1949 |
+
"grad_norm": 1.5173367261886597,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.4519,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 13.744930148715637,
|
| 1956 |
+
"eval_accuracy": 0.8938849251143673,
|
| 1957 |
+
"eval_loss": 0.44248726963996887,
|
| 1958 |
+
"eval_runtime": 220.2187,
|
| 1959 |
+
"eval_samples_per_second": 125.575,
|
| 1960 |
+
"eval_steps_per_second": 3.928,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 13.857593510590355,
|
| 1965 |
+
"grad_norm": 1.6886624097824097,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.4492,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 13.857593510590355,
|
| 1972 |
+
"eval_accuracy": 0.8941456344925675,
|
| 1973 |
+
"eval_loss": 0.44254130125045776,
|
| 1974 |
+
"eval_runtime": 220.9103,
|
| 1975 |
+
"eval_samples_per_second": 125.182,
|
| 1976 |
+
"eval_steps_per_second": 3.916,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 13.970256872465074,
|
| 1981 |
+
"grad_norm": 1.560421109199524,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.4495,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 13.970256872465074,
|
| 1988 |
+
"eval_accuracy": 0.8943100925877457,
|
| 1989 |
+
"eval_loss": 0.43967217206954956,
|
| 1990 |
+
"eval_runtime": 221.8474,
|
| 1991 |
+
"eval_samples_per_second": 124.653,
|
| 1992 |
+
"eval_steps_per_second": 3.899,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 14.082920234339793,
|
| 1997 |
+
"grad_norm": 2.146169662475586,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.4438,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 14.082920234339793,
|
| 2004 |
+
"eval_accuracy": 0.8951792684420733,
|
| 2005 |
+
"eval_loss": 0.43378108739852905,
|
| 2006 |
+
"eval_runtime": 221.3932,
|
| 2007 |
+
"eval_samples_per_second": 124.909,
|
| 2008 |
+
"eval_steps_per_second": 3.907,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 14.195583596214512,
|
| 2013 |
+
"grad_norm": 1.6352427005767822,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.4437,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 14.195583596214512,
|
| 2020 |
+
"eval_accuracy": 0.8944950022699799,
|
| 2021 |
+
"eval_loss": 0.4368970990180969,
|
| 2022 |
+
"eval_runtime": 221.6512,
|
| 2023 |
+
"eval_samples_per_second": 124.764,
|
| 2024 |
+
"eval_steps_per_second": 3.903,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 14.30824695808923,
|
| 2029 |
+
"grad_norm": 1.6570438146591187,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.4407,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 14.30824695808923,
|
| 2036 |
+
"eval_accuracy": 0.8954035156096067,
|
| 2037 |
+
"eval_loss": 0.43203291296958923,
|
| 2038 |
+
"eval_runtime": 220.9732,
|
| 2039 |
+
"eval_samples_per_second": 125.146,
|
| 2040 |
+
"eval_steps_per_second": 3.915,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 14.420910319963948,
|
| 2045 |
+
"grad_norm": 1.6666638851165771,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.4409,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 14.420910319963948,
|
| 2052 |
+
"eval_accuracy": 0.8951364676803809,
|
| 2053 |
+
"eval_loss": 0.43579936027526855,
|
| 2054 |
+
"eval_runtime": 221.9618,
|
| 2055 |
+
"eval_samples_per_second": 124.589,
|
| 2056 |
+
"eval_steps_per_second": 3.897,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 14.533573681838666,
|
| 2061 |
+
"grad_norm": 1.5540229082107544,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.4425,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 14.533573681838666,
|
| 2068 |
+
"eval_accuracy": 0.8954575425427099,
|
| 2069 |
+
"eval_loss": 0.432124525308609,
|
| 2070 |
+
"eval_runtime": 222.2861,
|
| 2071 |
+
"eval_samples_per_second": 124.407,
|
| 2072 |
+
"eval_steps_per_second": 3.891,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 14.646237043713384,
|
| 2077 |
+
"grad_norm": 1.6039586067199707,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.4375,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 14.646237043713384,
|
| 2084 |
+
"eval_accuracy": 0.8959293125901446,
|
| 2085 |
+
"eval_loss": 0.4307084083557129,
|
| 2086 |
+
"eval_runtime": 221.3855,
|
| 2087 |
+
"eval_samples_per_second": 124.913,
|
| 2088 |
+
"eval_steps_per_second": 3.907,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 14.758900405588102,
|
| 2093 |
+
"grad_norm": 1.4141193628311157,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.4412,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 14.758900405588102,
|
| 2100 |
+
"eval_accuracy": 0.8955789560165742,
|
| 2101 |
+
"eval_loss": 0.4335871934890747,
|
| 2102 |
+
"eval_runtime": 220.8948,
|
| 2103 |
+
"eval_samples_per_second": 125.191,
|
| 2104 |
+
"eval_steps_per_second": 3.916,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 14.87156376746282,
|
| 2109 |
+
"grad_norm": 1.479407548904419,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.4402,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 14.87156376746282,
|
| 2116 |
+
"eval_accuracy": 0.8960790351082009,
|
| 2117 |
+
"eval_loss": 0.428357869386673,
|
| 2118 |
+
"eval_runtime": 221.3572,
|
| 2119 |
+
"eval_samples_per_second": 124.929,
|
| 2120 |
+
"eval_steps_per_second": 3.908,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 14.984227129337539,
|
| 2125 |
+
"grad_norm": 1.6063992977142334,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.4386,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 14.984227129337539,
|
| 2132 |
+
"eval_accuracy": 0.8961887828028311,
|
| 2133 |
+
"eval_loss": 0.42679697275161743,
|
| 2134 |
+
"eval_runtime": 220.1773,
|
| 2135 |
+
"eval_samples_per_second": 125.599,
|
| 2136 |
+
"eval_steps_per_second": 3.929,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 15.096890491212259,
|
| 2141 |
+
"grad_norm": 1.7383469343185425,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.4342,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 15.096890491212259,
|
| 2148 |
+
"eval_accuracy": 0.8963356120392529,
|
| 2149 |
+
"eval_loss": 0.43058517575263977,
|
| 2150 |
+
"eval_runtime": 221.0267,
|
| 2151 |
+
"eval_samples_per_second": 125.116,
|
| 2152 |
+
"eval_steps_per_second": 3.914,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 15.209553853086977,
|
| 2157 |
+
"grad_norm": 1.4529184103012085,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.4355,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 15.209553853086977,
|
| 2164 |
+
"eval_accuracy": 0.8963537523060265,
|
| 2165 |
+
"eval_loss": 0.4319207966327667,
|
| 2166 |
+
"eval_runtime": 221.0822,
|
| 2167 |
+
"eval_samples_per_second": 125.085,
|
| 2168 |
+
"eval_steps_per_second": 3.913,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 15.322217214961695,
|
| 2173 |
+
"grad_norm": 1.5925979614257812,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.434,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 15.322217214961695,
|
| 2180 |
+
"eval_accuracy": 0.8967275614111444,
|
| 2181 |
+
"eval_loss": 0.4326106905937195,
|
| 2182 |
+
"eval_runtime": 221.1171,
|
| 2183 |
+
"eval_samples_per_second": 125.065,
|
| 2184 |
+
"eval_steps_per_second": 3.912,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 15.434880576836413,
|
| 2189 |
+
"grad_norm": 1.5591844320297241,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.4299,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 15.434880576836413,
|
| 2196 |
+
"eval_accuracy": 0.8968720289079662,
|
| 2197 |
+
"eval_loss": 0.42554253339767456,
|
| 2198 |
+
"eval_runtime": 221.0599,
|
| 2199 |
+
"eval_samples_per_second": 125.097,
|
| 2200 |
+
"eval_steps_per_second": 3.913,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 15.547543938711131,
|
| 2205 |
+
"grad_norm": 1.6964800357818604,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.4302,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 15.547543938711131,
|
| 2212 |
+
"eval_accuracy": 0.8968983814075581,
|
| 2213 |
+
"eval_loss": 0.43178391456604004,
|
| 2214 |
+
"eval_runtime": 220.2723,
|
| 2215 |
+
"eval_samples_per_second": 125.545,
|
| 2216 |
+
"eval_steps_per_second": 3.927,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 15.66020730058585,
|
| 2221 |
+
"grad_norm": 1.7176204919815063,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.4317,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 15.66020730058585,
|
| 2228 |
+
"eval_accuracy": 0.8971937797880897,
|
| 2229 |
+
"eval_loss": 0.42581045627593994,
|
| 2230 |
+
"eval_runtime": 221.117,
|
| 2231 |
+
"eval_samples_per_second": 125.065,
|
| 2232 |
+
"eval_steps_per_second": 3.912,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 15.772870662460567,
|
| 2237 |
+
"grad_norm": 1.4802976846694946,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.4335,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 15.772870662460567,
|
| 2244 |
+
"eval_accuracy": 0.8972389358742884,
|
| 2245 |
+
"eval_loss": 0.4227333068847656,
|
| 2246 |
+
"eval_runtime": 220.2787,
|
| 2247 |
+
"eval_samples_per_second": 125.541,
|
| 2248 |
+
"eval_steps_per_second": 3.927,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 15.885534024335286,
|
| 2253 |
+
"grad_norm": 1.4625871181488037,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.4313,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 15.885534024335286,
|
| 2260 |
+
"eval_accuracy": 0.8974343061715017,
|
| 2261 |
+
"eval_loss": 0.420085072517395,
|
| 2262 |
+
"eval_runtime": 221.0709,
|
| 2263 |
+
"eval_samples_per_second": 125.091,
|
| 2264 |
+
"eval_steps_per_second": 3.913,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 15.998197386210004,
|
| 2269 |
+
"grad_norm": 1.4574440717697144,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.4288,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 15.998197386210004,
|
| 2276 |
+
"eval_accuracy": 0.8976216192291354,
|
| 2277 |
+
"eval_loss": 0.42089083790779114,
|
| 2278 |
+
"eval_runtime": 221.0359,
|
| 2279 |
+
"eval_samples_per_second": 125.111,
|
| 2280 |
+
"eval_steps_per_second": 3.913,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 16.110860748084722,
|
| 2285 |
+
"grad_norm": 1.415560245513916,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.4245,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 16.110860748084722,
|
| 2292 |
+
"eval_accuracy": 0.8975563777935319,
|
| 2293 |
+
"eval_loss": 0.42667824029922485,
|
| 2294 |
+
"eval_runtime": 220.6503,
|
| 2295 |
+
"eval_samples_per_second": 125.33,
|
| 2296 |
+
"eval_steps_per_second": 3.92,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 16.223524109959442,
|
| 2301 |
+
"grad_norm": 1.6393336057662964,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.4271,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 16.223524109959442,
|
| 2308 |
+
"eval_accuracy": 0.8984646682572308,
|
| 2309 |
+
"eval_loss": 0.4213043749332428,
|
| 2310 |
+
"eval_runtime": 219.8857,
|
| 2311 |
+
"eval_samples_per_second": 125.765,
|
| 2312 |
+
"eval_steps_per_second": 3.934,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 16.336187471834158,
|
| 2317 |
+
"grad_norm": 1.6446831226348877,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.4234,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 16.336187471834158,
|
| 2324 |
+
"eval_accuracy": 0.8985838129375973,
|
| 2325 |
+
"eval_loss": 0.42193278670310974,
|
| 2326 |
+
"eval_runtime": 221.0608,
|
| 2327 |
+
"eval_samples_per_second": 125.097,
|
| 2328 |
+
"eval_steps_per_second": 3.913,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 16.448850833708878,
|
| 2333 |
+
"grad_norm": 1.725674033164978,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.4251,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 16.448850833708878,
|
| 2340 |
+
"eval_accuracy": 0.8987038252593214,
|
| 2341 |
+
"eval_loss": 0.4171987771987915,
|
| 2342 |
+
"eval_runtime": 220.8886,
|
| 2343 |
+
"eval_samples_per_second": 125.194,
|
| 2344 |
+
"eval_steps_per_second": 3.916,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 16.561514195583594,
|
| 2349 |
+
"grad_norm": 1.5979257822036743,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.4217,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 16.561514195583594,
|
| 2356 |
+
"eval_accuracy": 0.8987482542802462,
|
| 2357 |
+
"eval_loss": 0.418377161026001,
|
| 2358 |
+
"eval_runtime": 221.121,
|
| 2359 |
+
"eval_samples_per_second": 125.063,
|
| 2360 |
+
"eval_steps_per_second": 3.912,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 16.674177557458314,
|
| 2365 |
+
"grad_norm": 1.4892100095748901,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.4205,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 16.674177557458314,
|
| 2372 |
+
"eval_accuracy": 0.8989114915507669,
|
| 2373 |
+
"eval_loss": 0.4165091812610626,
|
| 2374 |
+
"eval_runtime": 222.5924,
|
| 2375 |
+
"eval_samples_per_second": 124.236,
|
| 2376 |
+
"eval_steps_per_second": 3.886,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 16.786840919333034,
|
| 2381 |
+
"grad_norm": 1.4461849927902222,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.4228,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 16.786840919333034,
|
| 2388 |
+
"eval_accuracy": 0.8989125542022752,
|
| 2389 |
+
"eval_loss": 0.4175247848033905,
|
| 2390 |
+
"eval_runtime": 222.3981,
|
| 2391 |
+
"eval_samples_per_second": 124.345,
|
| 2392 |
+
"eval_steps_per_second": 3.889,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 16.89950428120775,
|
| 2397 |
+
"grad_norm": 1.768370509147644,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.421,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 16.89950428120775,
|
| 2404 |
+
"eval_accuracy": 0.8991065894891168,
|
| 2405 |
+
"eval_loss": 0.41619065403938293,
|
| 2406 |
+
"eval_runtime": 222.6024,
|
| 2407 |
+
"eval_samples_per_second": 124.23,
|
| 2408 |
+
"eval_steps_per_second": 3.886,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 17.01216764308247,
|
| 2413 |
+
"grad_norm": 1.4250850677490234,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.4178,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 17.01216764308247,
|
| 2420 |
+
"eval_accuracy": 0.8994014895612595,
|
| 2421 |
+
"eval_loss": 0.4117368161678314,
|
| 2422 |
+
"eval_runtime": 222.14,
|
| 2423 |
+
"eval_samples_per_second": 124.489,
|
| 2424 |
+
"eval_steps_per_second": 3.894,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 17.124831004957187,
|
| 2429 |
+
"grad_norm": 1.4036965370178223,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.4176,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 17.124831004957187,
|
| 2436 |
+
"eval_accuracy": 0.8995830389073786,
|
| 2437 |
+
"eval_loss": 0.4121379852294922,
|
| 2438 |
+
"eval_runtime": 222.2839,
|
| 2439 |
+
"eval_samples_per_second": 124.408,
|
| 2440 |
+
"eval_steps_per_second": 3.891,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 17.237494366831907,
|
| 2445 |
+
"grad_norm": 1.395093321800232,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.4172,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 17.237494366831907,
|
| 2452 |
+
"eval_accuracy": 0.8998577766066815,
|
| 2453 |
+
"eval_loss": 0.41285398602485657,
|
| 2454 |
+
"eval_runtime": 222.2267,
|
| 2455 |
+
"eval_samples_per_second": 124.44,
|
| 2456 |
+
"eval_steps_per_second": 3.892,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 17.350157728706623,
|
| 2461 |
+
"grad_norm": 1.5492697954177856,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.4133,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 17.350157728706623,
|
| 2468 |
+
"eval_accuracy": 0.8992890854661668,
|
| 2469 |
+
"eval_loss": 0.41492369771003723,
|
| 2470 |
+
"eval_runtime": 219.0021,
|
| 2471 |
+
"eval_samples_per_second": 126.273,
|
| 2472 |
+
"eval_steps_per_second": 3.95,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 17.462821090581343,
|
| 2477 |
+
"grad_norm": 1.4863234758377075,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.4166,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 17.462821090581343,
|
| 2484 |
+
"eval_accuracy": 0.8995439142560963,
|
| 2485 |
+
"eval_loss": 0.41370296478271484,
|
| 2486 |
+
"eval_runtime": 220.4874,
|
| 2487 |
+
"eval_samples_per_second": 125.422,
|
| 2488 |
+
"eval_steps_per_second": 3.923,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 17.57548445245606,
|
| 2493 |
+
"grad_norm": 1.8134657144546509,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.4167,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 17.57548445245606,
|
| 2500 |
+
"eval_accuracy": 0.8998953243247179,
|
| 2501 |
+
"eval_loss": 0.4118014872074127,
|
| 2502 |
+
"eval_runtime": 220.8002,
|
| 2503 |
+
"eval_samples_per_second": 125.244,
|
| 2504 |
+
"eval_steps_per_second": 3.918,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 17.68814781433078,
|
| 2509 |
+
"grad_norm": 1.7903392314910889,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.4164,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 17.68814781433078,
|
| 2516 |
+
"eval_accuracy": 0.9001628949311502,
|
| 2517 |
+
"eval_loss": 0.4123002886772156,
|
| 2518 |
+
"eval_runtime": 219.63,
|
| 2519 |
+
"eval_samples_per_second": 125.912,
|
| 2520 |
+
"eval_steps_per_second": 3.938,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 17.8008111762055,
|
| 2525 |
+
"grad_norm": 1.6216607093811035,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.4143,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 17.8008111762055,
|
| 2532 |
+
"eval_accuracy": 0.9001079811521843,
|
| 2533 |
+
"eval_loss": 0.40997758507728577,
|
| 2534 |
+
"eval_runtime": 219.8198,
|
| 2535 |
+
"eval_samples_per_second": 125.803,
|
| 2536 |
+
"eval_steps_per_second": 3.935,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 17.913474538080216,
|
| 2541 |
+
"grad_norm": 1.5128173828125,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.4136,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 17.913474538080216,
|
| 2548 |
+
"eval_accuracy": 0.9006287821890727,
|
| 2549 |
+
"eval_loss": 0.41052308678627014,
|
| 2550 |
+
"eval_runtime": 219.697,
|
| 2551 |
+
"eval_samples_per_second": 125.873,
|
| 2552 |
+
"eval_steps_per_second": 3.937,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 18.026137899954936,
|
| 2557 |
+
"grad_norm": 1.413712978363037,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.4132,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 18.026137899954936,
|
| 2564 |
+
"eval_accuracy": 0.9007660373895346,
|
| 2565 |
+
"eval_loss": 0.4081571400165558,
|
| 2566 |
+
"eval_runtime": 220.6703,
|
| 2567 |
+
"eval_samples_per_second": 125.318,
|
| 2568 |
+
"eval_steps_per_second": 3.92,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 18.138801261829652,
|
| 2573 |
+
"grad_norm": 1.7320311069488525,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.4102,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 18.138801261829652,
|
| 2580 |
+
"eval_accuracy": 0.9009554825729237,
|
| 2581 |
+
"eval_loss": 0.407240092754364,
|
| 2582 |
+
"eval_runtime": 221.4224,
|
| 2583 |
+
"eval_samples_per_second": 124.893,
|
| 2584 |
+
"eval_steps_per_second": 3.907,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 18.251464623704372,
|
| 2589 |
+
"grad_norm": 1.8033103942871094,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.4097,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 18.251464623704372,
|
| 2596 |
+
"eval_accuracy": 0.9008985457774398,
|
| 2597 |
+
"eval_loss": 0.4109956920146942,
|
| 2598 |
+
"eval_runtime": 221.1014,
|
| 2599 |
+
"eval_samples_per_second": 125.074,
|
| 2600 |
+
"eval_steps_per_second": 3.912,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 18.36412798557909,
|
| 2605 |
+
"grad_norm": 1.8222883939743042,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.4085,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 18.36412798557909,
|
| 2612 |
+
"eval_accuracy": 0.9007539025464132,
|
| 2613 |
+
"eval_loss": 0.4095366299152374,
|
| 2614 |
+
"eval_runtime": 220.8203,
|
| 2615 |
+
"eval_samples_per_second": 125.233,
|
| 2616 |
+
"eval_steps_per_second": 3.917,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 18.47679134745381,
|
| 2621 |
+
"grad_norm": 1.4663125276565552,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.4105,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 18.47679134745381,
|
| 2628 |
+
"eval_accuracy": 0.9014532811520996,
|
| 2629 |
+
"eval_loss": 0.4047625958919525,
|
| 2630 |
+
"eval_runtime": 219.6263,
|
| 2631 |
+
"eval_samples_per_second": 125.914,
|
| 2632 |
+
"eval_steps_per_second": 3.939,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 18.589454709328525,
|
| 2637 |
+
"grad_norm": 1.8482975959777832,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.4096,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 18.589454709328525,
|
| 2644 |
+
"eval_accuracy": 0.9010233806097327,
|
| 2645 |
+
"eval_loss": 0.4072835445404053,
|
| 2646 |
+
"eval_runtime": 220.7586,
|
| 2647 |
+
"eval_samples_per_second": 125.268,
|
| 2648 |
+
"eval_steps_per_second": 3.918,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 18.702118071203245,
|
| 2653 |
+
"grad_norm": 1.4483723640441895,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.4041,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 18.702118071203245,
|
| 2660 |
+
"eval_accuracy": 0.9015295597674521,
|
| 2661 |
+
"eval_loss": 0.4039141833782196,
|
| 2662 |
+
"eval_runtime": 220.7392,
|
| 2663 |
+
"eval_samples_per_second": 125.279,
|
| 2664 |
+
"eval_steps_per_second": 3.919,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 18.814781433077965,
|
| 2669 |
+
"grad_norm": 1.6040253639221191,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.4062,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 18.814781433077965,
|
| 2676 |
+
"eval_accuracy": 0.9016612318058135,
|
| 2677 |
+
"eval_loss": 0.40488725900650024,
|
| 2678 |
+
"eval_runtime": 221.3884,
|
| 2679 |
+
"eval_samples_per_second": 124.912,
|
| 2680 |
+
"eval_steps_per_second": 3.907,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 18.92744479495268,
|
| 2685 |
+
"grad_norm": 1.3560248613357544,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.4045,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 18.92744479495268,
|
| 2692 |
+
"eval_accuracy": 0.9015874866980568,
|
| 2693 |
+
"eval_loss": 0.4032597243785858,
|
| 2694 |
+
"eval_runtime": 221.9037,
|
| 2695 |
+
"eval_samples_per_second": 124.622,
|
| 2696 |
+
"eval_steps_per_second": 3.898,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 19.0401081568274,
|
| 2701 |
+
"grad_norm": 1.6236895322799683,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.4038,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 19.0401081568274,
|
| 2708 |
+
"eval_accuracy": 0.901710217516976,
|
| 2709 |
+
"eval_loss": 0.4084183871746063,
|
| 2710 |
+
"eval_runtime": 220.8431,
|
| 2711 |
+
"eval_samples_per_second": 125.22,
|
| 2712 |
+
"eval_steps_per_second": 3.917,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 19.152771518702117,
|
| 2717 |
+
"grad_norm": 1.6514983177185059,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.4037,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 19.152771518702117,
|
| 2724 |
+
"eval_accuracy": 0.9016946022320732,
|
| 2725 |
+
"eval_loss": 0.4033704102039337,
|
| 2726 |
+
"eval_runtime": 221.6212,
|
| 2727 |
+
"eval_samples_per_second": 124.78,
|
| 2728 |
+
"eval_steps_per_second": 3.903,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 19.265434880576837,
|
| 2733 |
+
"grad_norm": 1.3684407472610474,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.4022,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 19.265434880576837,
|
| 2740 |
+
"eval_accuracy": 0.9021324676993308,
|
| 2741 |
+
"eval_loss": 0.40617531538009644,
|
| 2742 |
+
"eval_runtime": 221.8256,
|
| 2743 |
+
"eval_samples_per_second": 124.666,
|
| 2744 |
+
"eval_steps_per_second": 3.899,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 19.378098242451554,
|
| 2749 |
+
"grad_norm": 1.592301607131958,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.4059,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 19.378098242451554,
|
| 2756 |
+
"eval_accuracy": 0.902363044454423,
|
| 2757 |
+
"eval_loss": 0.3991073668003082,
|
| 2758 |
+
"eval_runtime": 220.8011,
|
| 2759 |
+
"eval_samples_per_second": 125.244,
|
| 2760 |
+
"eval_steps_per_second": 3.918,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 19.490761604326273,
|
| 2765 |
+
"grad_norm": 1.5463926792144775,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.4013,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 19.490761604326273,
|
| 2772 |
+
"eval_accuracy": 0.9023868906868481,
|
| 2773 |
+
"eval_loss": 0.39859089255332947,
|
| 2774 |
+
"eval_runtime": 220.6504,
|
| 2775 |
+
"eval_samples_per_second": 125.329,
|
| 2776 |
+
"eval_steps_per_second": 3.92,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 19.60342496620099,
|
| 2781 |
+
"grad_norm": 1.6952037811279297,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.4004,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 19.60342496620099,
|
| 2788 |
+
"eval_accuracy": 0.9029012333672634,
|
| 2789 |
+
"eval_loss": 0.4017859995365143,
|
| 2790 |
+
"eval_runtime": 220.6857,
|
| 2791 |
+
"eval_samples_per_second": 125.309,
|
| 2792 |
+
"eval_steps_per_second": 3.92,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 19.71608832807571,
|
| 2797 |
+
"grad_norm": 1.5156389474868774,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.4023,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 19.71608832807571,
|
| 2804 |
+
"eval_accuracy": 0.9022691715502759,
|
| 2805 |
+
"eval_loss": 0.40082216262817383,
|
| 2806 |
+
"eval_runtime": 220.7786,
|
| 2807 |
+
"eval_samples_per_second": 125.257,
|
| 2808 |
+
"eval_steps_per_second": 3.918,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 19.82875168995043,
|
| 2813 |
+
"grad_norm": 1.5951709747314453,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.3987,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 19.82875168995043,
|
| 2820 |
+
"eval_accuracy": 0.9028266490406112,
|
| 2821 |
+
"eval_loss": 0.4010894000530243,
|
| 2822 |
+
"eval_runtime": 220.1664,
|
| 2823 |
+
"eval_samples_per_second": 125.605,
|
| 2824 |
+
"eval_steps_per_second": 3.929,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 19.941415051825146,
|
| 2829 |
+
"grad_norm": 1.4990533590316772,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.3935,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 19.941415051825146,
|
| 2836 |
+
"eval_accuracy": 0.9027395900326748,
|
| 2837 |
+
"eval_loss": 0.401162326335907,
|
| 2838 |
+
"eval_runtime": 220.111,
|
| 2839 |
+
"eval_samples_per_second": 125.637,
|
| 2840 |
+
"eval_steps_per_second": 3.93,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 20.054078413699866,
|
| 2845 |
+
"grad_norm": 1.5961695909500122,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.3978,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 20.054078413699866,
|
| 2852 |
+
"eval_accuracy": 0.902977115716753,
|
| 2853 |
+
"eval_loss": 0.3981638252735138,
|
| 2854 |
+
"eval_runtime": 219.6972,
|
| 2855 |
+
"eval_samples_per_second": 125.873,
|
| 2856 |
+
"eval_steps_per_second": 3.937,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 20.166741775574582,
|
| 2861 |
+
"grad_norm": 1.5186184644699097,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.4012,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 20.166741775574582,
|
| 2868 |
+
"eval_accuracy": 0.9029895131243953,
|
| 2869 |
+
"eval_loss": 0.39535069465637207,
|
| 2870 |
+
"eval_runtime": 220.4034,
|
| 2871 |
+
"eval_samples_per_second": 125.47,
|
| 2872 |
+
"eval_steps_per_second": 3.925,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 20.279405137449302,
|
| 2877 |
+
"grad_norm": 1.7340284585952759,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.3954,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 20.279405137449302,
|
| 2884 |
+
"eval_accuracy": 0.9031845731573412,
|
| 2885 |
+
"eval_loss": 0.3972371816635132,
|
| 2886 |
+
"eval_runtime": 220.9175,
|
| 2887 |
+
"eval_samples_per_second": 125.178,
|
| 2888 |
+
"eval_steps_per_second": 3.915,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 20.39206849932402,
|
| 2893 |
+
"grad_norm": 1.4601465463638306,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.3933,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 20.39206849932402,
|
| 2900 |
+
"eval_accuracy": 0.9031702530935091,
|
| 2901 |
+
"eval_loss": 0.39749225974082947,
|
| 2902 |
+
"eval_runtime": 220.5834,
|
| 2903 |
+
"eval_samples_per_second": 125.368,
|
| 2904 |
+
"eval_steps_per_second": 3.921,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 20.50473186119874,
|
| 2909 |
+
"grad_norm": 1.6822484731674194,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.3985,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 20.50473186119874,
|
| 2916 |
+
"eval_accuracy": 0.903283638473266,
|
| 2917 |
+
"eval_loss": 0.39412999153137207,
|
| 2918 |
+
"eval_runtime": 220.402,
|
| 2919 |
+
"eval_samples_per_second": 125.471,
|
| 2920 |
+
"eval_steps_per_second": 3.925,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 20.617395223073455,
|
| 2925 |
+
"grad_norm": 1.5493133068084717,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.3952,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 20.617395223073455,
|
| 2932 |
+
"eval_accuracy": 0.9031870870760611,
|
| 2933 |
+
"eval_loss": 0.39998504519462585,
|
| 2934 |
+
"eval_runtime": 219.1703,
|
| 2935 |
+
"eval_samples_per_second": 126.176,
|
| 2936 |
+
"eval_steps_per_second": 3.947,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 20.730058584948175,
|
| 2941 |
+
"grad_norm": 1.6142163276672363,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.395,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 20.730058584948175,
|
| 2948 |
+
"eval_accuracy": 0.9037042508521438,
|
| 2949 |
+
"eval_loss": 0.39454683661460876,
|
| 2950 |
+
"eval_runtime": 220.1482,
|
| 2951 |
+
"eval_samples_per_second": 125.615,
|
| 2952 |
+
"eval_steps_per_second": 3.929,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 20.842721946822895,
|
| 2957 |
+
"grad_norm": 1.3768945932388306,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.3925,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 20.842721946822895,
|
| 2964 |
+
"eval_accuracy": 0.9035520393735632,
|
| 2965 |
+
"eval_loss": 0.3969292640686035,
|
| 2966 |
+
"eval_runtime": 218.8787,
|
| 2967 |
+
"eval_samples_per_second": 126.344,
|
| 2968 |
+
"eval_steps_per_second": 3.952,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 20.95538530869761,
|
| 2973 |
+
"grad_norm": 1.8161870241165161,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.3911,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 20.95538530869761,
|
| 2980 |
+
"eval_accuracy": 0.9034115695768419,
|
| 2981 |
+
"eval_loss": 0.39153432846069336,
|
| 2982 |
+
"eval_runtime": 219.6974,
|
| 2983 |
+
"eval_samples_per_second": 125.873,
|
| 2984 |
+
"eval_steps_per_second": 3.937,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 21.06804867057233,
|
| 2989 |
+
"grad_norm": 1.7550774812698364,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.3927,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 21.06804867057233,
|
| 2996 |
+
"eval_accuracy": 0.9035121668560334,
|
| 2997 |
+
"eval_loss": 0.39775171875953674,
|
| 2998 |
+
"eval_runtime": 221.0095,
|
| 2999 |
+
"eval_samples_per_second": 125.126,
|
| 3000 |
+
"eval_steps_per_second": 3.914,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 21.180712032447047,
|
| 3005 |
+
"grad_norm": 1.5582369565963745,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.3891,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 21.180712032447047,
|
| 3012 |
+
"eval_accuracy": 0.9037201879273532,
|
| 3013 |
+
"eval_loss": 0.3943246006965637,
|
| 3014 |
+
"eval_runtime": 220.3117,
|
| 3015 |
+
"eval_samples_per_second": 125.522,
|
| 3016 |
+
"eval_steps_per_second": 3.926,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 21.293375394321767,
|
| 3021 |
+
"grad_norm": 1.6729559898376465,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.3912,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 21.293375394321767,
|
| 3028 |
+
"eval_accuracy": 0.9036670141570837,
|
| 3029 |
+
"eval_loss": 0.39444249868392944,
|
| 3030 |
+
"eval_runtime": 219.9471,
|
| 3031 |
+
"eval_samples_per_second": 125.73,
|
| 3032 |
+
"eval_steps_per_second": 3.933,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 21.406038756196484,
|
| 3037 |
+
"grad_norm": 1.6871699094772339,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.3908,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 21.406038756196484,
|
| 3044 |
+
"eval_accuracy": 0.9037736297217607,
|
| 3045 |
+
"eval_loss": 0.39369192719459534,
|
| 3046 |
+
"eval_runtime": 219.9205,
|
| 3047 |
+
"eval_samples_per_second": 125.745,
|
| 3048 |
+
"eval_steps_per_second": 3.933,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 21.518702118071204,
|
| 3053 |
+
"grad_norm": 1.486741304397583,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.3902,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 21.518702118071204,
|
| 3060 |
+
"eval_accuracy": 0.9034302972672164,
|
| 3061 |
+
"eval_loss": 0.39573636651039124,
|
| 3062 |
+
"eval_runtime": 219.8759,
|
| 3063 |
+
"eval_samples_per_second": 125.771,
|
| 3064 |
+
"eval_steps_per_second": 3.934,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 21.63136547994592,
|
| 3069 |
+
"grad_norm": 1.8056081533432007,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.3891,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 21.63136547994592,
|
| 3076 |
+
"eval_accuracy": 0.9045647365783699,
|
| 3077 |
+
"eval_loss": 0.39023157954216003,
|
| 3078 |
+
"eval_runtime": 221.3034,
|
| 3079 |
+
"eval_samples_per_second": 124.96,
|
| 3080 |
+
"eval_steps_per_second": 3.909,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 21.74402884182064,
|
| 3085 |
+
"grad_norm": 1.552370309829712,
|
| 3086 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 3087 |
+
"loss": 0.3894,
|
| 3088 |
+
"step": 96500
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 21.74402884182064,
|
| 3092 |
+
"eval_accuracy": 0.9044615558398447,
|
| 3093 |
+
"eval_loss": 0.39400991797447205,
|
| 3094 |
+
"eval_runtime": 219.8746,
|
| 3095 |
+
"eval_samples_per_second": 125.772,
|
| 3096 |
+
"eval_steps_per_second": 3.934,
|
| 3097 |
+
"step": 96500
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 21.85669220369536,
|
| 3101 |
+
"grad_norm": 1.506536841392517,
|
| 3102 |
+
"learning_rate": 3e-06,
|
| 3103 |
+
"loss": 0.3904,
|
| 3104 |
+
"step": 97000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 21.85669220369536,
|
| 3108 |
+
"eval_accuracy": 0.9044962394479266,
|
| 3109 |
+
"eval_loss": 0.390458881855011,
|
| 3110 |
+
"eval_runtime": 220.131,
|
| 3111 |
+
"eval_samples_per_second": 125.625,
|
| 3112 |
+
"eval_steps_per_second": 3.929,
|
| 3113 |
+
"step": 97000
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 21.969355565570076,
|
| 3117 |
+
"grad_norm": 1.6080279350280762,
|
| 3118 |
+
"learning_rate": 2.5e-06,
|
| 3119 |
+
"loss": 0.3882,
|
| 3120 |
+
"step": 97500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 21.969355565570076,
|
| 3124 |
+
"eval_accuracy": 0.9043700852475594,
|
| 3125 |
+
"eval_loss": 0.39395132660865784,
|
| 3126 |
+
"eval_runtime": 220.1175,
|
| 3127 |
+
"eval_samples_per_second": 125.633,
|
| 3128 |
+
"eval_steps_per_second": 3.93,
|
| 3129 |
+
"step": 97500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 22.082018927444796,
|
| 3133 |
+
"grad_norm": 1.6551542282104492,
|
| 3134 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 3135 |
+
"loss": 0.388,
|
| 3136 |
+
"step": 98000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 22.082018927444796,
|
| 3140 |
+
"eval_accuracy": 0.904642958920198,
|
| 3141 |
+
"eval_loss": 0.39477479457855225,
|
| 3142 |
+
"eval_runtime": 219.108,
|
| 3143 |
+
"eval_samples_per_second": 126.212,
|
| 3144 |
+
"eval_steps_per_second": 3.948,
|
| 3145 |
+
"step": 98000
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 22.194682289319513,
|
| 3149 |
+
"grad_norm": 1.3376331329345703,
|
| 3150 |
+
"learning_rate": 1.5e-06,
|
| 3151 |
+
"loss": 0.3888,
|
| 3152 |
+
"step": 98500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 22.194682289319513,
|
| 3156 |
+
"eval_accuracy": 0.9042594879589607,
|
| 3157 |
+
"eval_loss": 0.39155128598213196,
|
| 3158 |
+
"eval_runtime": 221.2476,
|
| 3159 |
+
"eval_samples_per_second": 124.991,
|
| 3160 |
+
"eval_steps_per_second": 3.91,
|
| 3161 |
+
"step": 98500
|
| 3162 |
+
},
|
| 3163 |
+
{
|
| 3164 |
+
"epoch": 22.307345651194233,
|
| 3165 |
+
"grad_norm": 1.6391901969909668,
|
| 3166 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 3167 |
+
"loss": 0.385,
|
| 3168 |
+
"step": 99000
|
| 3169 |
+
},
|
| 3170 |
+
{
|
| 3171 |
+
"epoch": 22.307345651194233,
|
| 3172 |
+
"eval_accuracy": 0.9047423169505552,
|
| 3173 |
+
"eval_loss": 0.3867943286895752,
|
| 3174 |
+
"eval_runtime": 220.9463,
|
| 3175 |
+
"eval_samples_per_second": 125.162,
|
| 3176 |
+
"eval_steps_per_second": 3.915,
|
| 3177 |
+
"step": 99000
|
| 3178 |
+
}
|
| 3179 |
+
],
|
| 3180 |
+
"logging_steps": 500,
|
| 3181 |
+
"max_steps": 100000,
|
| 3182 |
+
"num_input_tokens_seen": 0,
|
| 3183 |
+
"num_train_epochs": 23,
|
| 3184 |
+
"save_steps": 500,
|
| 3185 |
+
"stateful_callbacks": {
|
| 3186 |
+
"TrainerControl": {
|
| 3187 |
+
"args": {
|
| 3188 |
+
"should_epoch_stop": false,
|
| 3189 |
+
"should_evaluate": false,
|
| 3190 |
+
"should_log": false,
|
| 3191 |
+
"should_save": true,
|
| 3192 |
+
"should_training_stop": false
|
| 3193 |
+
},
|
| 3194 |
+
"attributes": {}
|
| 3195 |
+
}
|
| 3196 |
+
},
|
| 3197 |
+
"total_flos": 8.346992290195046e+17,
|
| 3198 |
+
"train_batch_size": 16,
|
| 3199 |
+
"trial_name": null,
|
| 3200 |
+
"trial_params": null
|
| 3201 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:857aedc013b47e10c1aee25297baaa247a1a19ce6fd5f7dfee955439aecdfcb8
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|