Uploading checkpoint-99500 for xlm-r - kat-geor
Browse files- .gitattributes +1 -0
- README.md +67 -0
- config.json +34 -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
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +3 -0
- tokenizer_config.json +54 -0
- trainer_state.json +3217 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: FacebookAI/xlm-roberta-base
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: kat-Geor
|
| 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 |
+
# kat-Geor
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [FacebookAI/xlm-roberta-base](https://huggingface.co/FacebookAI/xlm-roberta-base) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.8029
|
| 22 |
+
- Accuracy: 0.8658
|
| 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 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "FacebookAI/xlm-roberta-base",
|
| 3 |
+
"adapters": {
|
| 4 |
+
"adapters": {},
|
| 5 |
+
"config_map": {},
|
| 6 |
+
"fusion_config_map": {},
|
| 7 |
+
"fusions": {}
|
| 8 |
+
},
|
| 9 |
+
"architectures": [
|
| 10 |
+
"XLMRobertaForMaskedLM"
|
| 11 |
+
],
|
| 12 |
+
"attention_probs_dropout_prob": 0.1,
|
| 13 |
+
"bos_token_id": 0,
|
| 14 |
+
"classifier_dropout": null,
|
| 15 |
+
"eos_token_id": 2,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.1,
|
| 18 |
+
"hidden_size": 768,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"layer_norm_eps": 1e-05,
|
| 22 |
+
"max_position_embeddings": 514,
|
| 23 |
+
"model_type": "xlm-roberta",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_hidden_layers": 12,
|
| 26 |
+
"output_past": true,
|
| 27 |
+
"pad_token_id": 1,
|
| 28 |
+
"position_embedding_type": "absolute",
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.45.2",
|
| 31 |
+
"type_vocab_size": 1,
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 250002
|
| 34 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1ec25cfd3f17e77d08057c939acaa3cd6eba6dc93d2a7ac7225263e45069119
|
| 3 |
+
size 1113205088
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e5b891fdfa0f197e5c3c0943d61baacf586abad1adf144f643161424311a1e3
|
| 3 |
+
size 2226530490
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4724debe8091f5456289a5afe20611e98d8a281d7013cc02934d00b1044cd66d
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7385f6ac1b46993c044977d21c817e3a9c0a2b9b52d56a76f26d617e4ea251f5
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c251be062ddaedd9f747377e845d2255c93418d986e7d7d4faf837f7e6411691
|
| 3 |
+
size 1064
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8373f9cd3d27591e1924426bcc1c8799bc5a9affc4fc857982c5d66668dd1f41
|
| 3 |
+
size 17082832
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"mask_token": "<mask>",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "<pad>",
|
| 51 |
+
"sep_token": "</s>",
|
| 52 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 53 |
+
"unk_token": "<unk>"
|
| 54 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3217 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7950243949890137,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/xlm-r/kat-Geor/checkpoint-99500",
|
| 4 |
+
"epoch": 32.34720416124838,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 99500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1625487646293888,
|
| 13 |
+
"grad_norm": 4.931751251220703,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.3006,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1625487646293888,
|
| 20 |
+
"eval_accuracy": 0.7623556146386671,
|
| 21 |
+
"eval_loss": 1.3419054746627808,
|
| 22 |
+
"eval_runtime": 224.5592,
|
| 23 |
+
"eval_samples_per_second": 90.996,
|
| 24 |
+
"eval_steps_per_second": 2.846,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.3250975292587776,
|
| 29 |
+
"grad_norm": 4.145026683807373,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.2154,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.3250975292587776,
|
| 36 |
+
"eval_accuracy": 0.7721005611086652,
|
| 37 |
+
"eval_loss": 1.2983847856521606,
|
| 38 |
+
"eval_runtime": 225.9654,
|
| 39 |
+
"eval_samples_per_second": 90.43,
|
| 40 |
+
"eval_steps_per_second": 2.828,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.48764629388816644,
|
| 45 |
+
"grad_norm": 4.589369297027588,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.1779,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.48764629388816644,
|
| 52 |
+
"eval_accuracy": 0.7790585887983502,
|
| 53 |
+
"eval_loss": 1.2703238725662231,
|
| 54 |
+
"eval_runtime": 226.7078,
|
| 55 |
+
"eval_samples_per_second": 90.134,
|
| 56 |
+
"eval_steps_per_second": 2.819,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6501950585175552,
|
| 61 |
+
"grad_norm": 4.552799701690674,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 1.1546,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.6501950585175552,
|
| 68 |
+
"eval_accuracy": 0.7819611016961427,
|
| 69 |
+
"eval_loss": 1.2609333992004395,
|
| 70 |
+
"eval_runtime": 226.665,
|
| 71 |
+
"eval_samples_per_second": 90.151,
|
| 72 |
+
"eval_steps_per_second": 2.819,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.812743823146944,
|
| 77 |
+
"grad_norm": 3.709968328475952,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 1.1259,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.812743823146944,
|
| 84 |
+
"eval_accuracy": 0.7877356734428504,
|
| 85 |
+
"eval_loss": 1.2131903171539307,
|
| 86 |
+
"eval_runtime": 226.8657,
|
| 87 |
+
"eval_samples_per_second": 90.071,
|
| 88 |
+
"eval_steps_per_second": 2.817,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.9752925877763329,
|
| 93 |
+
"grad_norm": 3.6233718395233154,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 1.0938,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.9752925877763329,
|
| 100 |
+
"eval_accuracy": 0.789409001949858,
|
| 101 |
+
"eval_loss": 1.2194353342056274,
|
| 102 |
+
"eval_runtime": 226.5843,
|
| 103 |
+
"eval_samples_per_second": 90.183,
|
| 104 |
+
"eval_steps_per_second": 2.82,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.1378413524057218,
|
| 109 |
+
"grad_norm": 3.7706315517425537,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 1.0728,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.1378413524057218,
|
| 116 |
+
"eval_accuracy": 0.7930909730559011,
|
| 117 |
+
"eval_loss": 1.2004605531692505,
|
| 118 |
+
"eval_runtime": 225.3386,
|
| 119 |
+
"eval_samples_per_second": 90.681,
|
| 120 |
+
"eval_steps_per_second": 2.836,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.3003901170351106,
|
| 125 |
+
"grad_norm": 3.654762029647827,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 1.067,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.3003901170351106,
|
| 132 |
+
"eval_accuracy": 0.7911667822625772,
|
| 133 |
+
"eval_loss": 1.206479549407959,
|
| 134 |
+
"eval_runtime": 229.9442,
|
| 135 |
+
"eval_samples_per_second": 88.865,
|
| 136 |
+
"eval_steps_per_second": 2.779,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.4629388816644995,
|
| 141 |
+
"grad_norm": 3.6047844886779785,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 1.0623,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.4629388816644995,
|
| 148 |
+
"eval_accuracy": 0.7950484617863337,
|
| 149 |
+
"eval_loss": 1.186224341392517,
|
| 150 |
+
"eval_runtime": 229.5346,
|
| 151 |
+
"eval_samples_per_second": 89.024,
|
| 152 |
+
"eval_steps_per_second": 2.784,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.6254876462938883,
|
| 157 |
+
"grad_norm": 3.174414873123169,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 1.0452,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.6254876462938883,
|
| 164 |
+
"eval_accuracy": 0.7986192006290228,
|
| 165 |
+
"eval_loss": 1.1624418497085571,
|
| 166 |
+
"eval_runtime": 227.8114,
|
| 167 |
+
"eval_samples_per_second": 89.697,
|
| 168 |
+
"eval_steps_per_second": 2.805,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.7880364109232771,
|
| 173 |
+
"grad_norm": 3.455960988998413,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 1.0203,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.7880364109232771,
|
| 180 |
+
"eval_accuracy": 0.7996825782405662,
|
| 181 |
+
"eval_loss": 1.160246729850769,
|
| 182 |
+
"eval_runtime": 227.8469,
|
| 183 |
+
"eval_samples_per_second": 89.683,
|
| 184 |
+
"eval_steps_per_second": 2.805,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.9505851755526658,
|
| 189 |
+
"grad_norm": 3.4478368759155273,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 1.0102,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.9505851755526658,
|
| 196 |
+
"eval_accuracy": 0.8028851426121806,
|
| 197 |
+
"eval_loss": 1.1382070779800415,
|
| 198 |
+
"eval_runtime": 225.8399,
|
| 199 |
+
"eval_samples_per_second": 90.48,
|
| 200 |
+
"eval_steps_per_second": 2.829,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.113133940182055,
|
| 205 |
+
"grad_norm": 3.5076828002929688,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 1.0007,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.113133940182055,
|
| 212 |
+
"eval_accuracy": 0.8036015201468353,
|
| 213 |
+
"eval_loss": 1.1308486461639404,
|
| 214 |
+
"eval_runtime": 227.3542,
|
| 215 |
+
"eval_samples_per_second": 89.877,
|
| 216 |
+
"eval_steps_per_second": 2.811,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.2756827048114436,
|
| 221 |
+
"grad_norm": 2.954378366470337,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 0.9828,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.2756827048114436,
|
| 228 |
+
"eval_accuracy": 0.80609597689472,
|
| 229 |
+
"eval_loss": 1.1300016641616821,
|
| 230 |
+
"eval_runtime": 226.3258,
|
| 231 |
+
"eval_samples_per_second": 90.286,
|
| 232 |
+
"eval_steps_per_second": 2.823,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.438231469440832,
|
| 237 |
+
"grad_norm": 3.6220898628234863,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 0.9738,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.438231469440832,
|
| 244 |
+
"eval_accuracy": 0.807549206838832,
|
| 245 |
+
"eval_loss": 1.1173433065414429,
|
| 246 |
+
"eval_runtime": 230.0766,
|
| 247 |
+
"eval_samples_per_second": 88.814,
|
| 248 |
+
"eval_steps_per_second": 2.777,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.6007802340702213,
|
| 253 |
+
"grad_norm": 3.0354886054992676,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 0.9726,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.6007802340702213,
|
| 260 |
+
"eval_accuracy": 0.8092077277721563,
|
| 261 |
+
"eval_loss": 1.1192331314086914,
|
| 262 |
+
"eval_runtime": 228.747,
|
| 263 |
+
"eval_samples_per_second": 89.33,
|
| 264 |
+
"eval_steps_per_second": 2.793,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.7633289986996097,
|
| 269 |
+
"grad_norm": 2.905066967010498,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 0.9639,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.7633289986996097,
|
| 276 |
+
"eval_accuracy": 0.8105359765383991,
|
| 277 |
+
"eval_loss": 1.0964252948760986,
|
| 278 |
+
"eval_runtime": 239.0977,
|
| 279 |
+
"eval_samples_per_second": 85.463,
|
| 280 |
+
"eval_steps_per_second": 2.673,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.925877763328999,
|
| 285 |
+
"grad_norm": 2.5866756439208984,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 0.9562,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.925877763328999,
|
| 292 |
+
"eval_accuracy": 0.8124049786000702,
|
| 293 |
+
"eval_loss": 1.0979543924331665,
|
| 294 |
+
"eval_runtime": 228.0298,
|
| 295 |
+
"eval_samples_per_second": 89.611,
|
| 296 |
+
"eval_steps_per_second": 2.802,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0884265279583873,
|
| 301 |
+
"grad_norm": 3.0912091732025146,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 0.9412,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.0884265279583873,
|
| 308 |
+
"eval_accuracy": 0.8128664613154557,
|
| 309 |
+
"eval_loss": 1.098233699798584,
|
| 310 |
+
"eval_runtime": 226.3594,
|
| 311 |
+
"eval_samples_per_second": 90.272,
|
| 312 |
+
"eval_steps_per_second": 2.823,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.250975292587776,
|
| 317 |
+
"grad_norm": 3.310755968093872,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 0.9404,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.250975292587776,
|
| 324 |
+
"eval_accuracy": 0.8141734154155298,
|
| 325 |
+
"eval_loss": 1.0834821462631226,
|
| 326 |
+
"eval_runtime": 229.2889,
|
| 327 |
+
"eval_samples_per_second": 89.119,
|
| 328 |
+
"eval_steps_per_second": 2.787,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.413524057217165,
|
| 333 |
+
"grad_norm": 3.2035183906555176,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 0.9345,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.413524057217165,
|
| 340 |
+
"eval_accuracy": 0.8148430925920552,
|
| 341 |
+
"eval_loss": 1.071053385734558,
|
| 342 |
+
"eval_runtime": 228.6307,
|
| 343 |
+
"eval_samples_per_second": 89.376,
|
| 344 |
+
"eval_steps_per_second": 2.795,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.576072821846554,
|
| 349 |
+
"grad_norm": 3.3350119590759277,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 0.9287,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.576072821846554,
|
| 356 |
+
"eval_accuracy": 0.8148937367897325,
|
| 357 |
+
"eval_loss": 1.0869466066360474,
|
| 358 |
+
"eval_runtime": 229.0653,
|
| 359 |
+
"eval_samples_per_second": 89.206,
|
| 360 |
+
"eval_steps_per_second": 2.79,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 3.7386215864759427,
|
| 365 |
+
"grad_norm": 3.0126075744628906,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 0.916,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.7386215864759427,
|
| 372 |
+
"eval_accuracy": 0.8164226040931482,
|
| 373 |
+
"eval_loss": 1.0795150995254517,
|
| 374 |
+
"eval_runtime": 227.3062,
|
| 375 |
+
"eval_samples_per_second": 89.896,
|
| 376 |
+
"eval_steps_per_second": 2.811,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.9011703511053315,
|
| 381 |
+
"grad_norm": 2.9025819301605225,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 0.9114,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.9011703511053315,
|
| 388 |
+
"eval_accuracy": 0.8172531893844662,
|
| 389 |
+
"eval_loss": 1.0739566087722778,
|
| 390 |
+
"eval_runtime": 227.5695,
|
| 391 |
+
"eval_samples_per_second": 89.792,
|
| 392 |
+
"eval_steps_per_second": 2.808,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.06371911573472,
|
| 397 |
+
"grad_norm": 3.1317036151885986,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 0.9104,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.06371911573472,
|
| 404 |
+
"eval_accuracy": 0.8181712807844302,
|
| 405 |
+
"eval_loss": 1.057742953300476,
|
| 406 |
+
"eval_runtime": 226.8271,
|
| 407 |
+
"eval_samples_per_second": 90.086,
|
| 408 |
+
"eval_steps_per_second": 2.817,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.22626788036411,
|
| 413 |
+
"grad_norm": 2.9879837036132812,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 0.8932,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.22626788036411,
|
| 420 |
+
"eval_accuracy": 0.819528608544276,
|
| 421 |
+
"eval_loss": 1.0524249076843262,
|
| 422 |
+
"eval_runtime": 227.5673,
|
| 423 |
+
"eval_samples_per_second": 89.793,
|
| 424 |
+
"eval_steps_per_second": 2.808,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 4.388816644993498,
|
| 429 |
+
"grad_norm": 3.1617486476898193,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 0.8884,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 4.388816644993498,
|
| 436 |
+
"eval_accuracy": 0.8202382159332025,
|
| 437 |
+
"eval_loss": 1.0618681907653809,
|
| 438 |
+
"eval_runtime": 228.7279,
|
| 439 |
+
"eval_samples_per_second": 89.338,
|
| 440 |
+
"eval_steps_per_second": 2.794,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.551365409622887,
|
| 445 |
+
"grad_norm": 3.287351131439209,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.8932,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.551365409622887,
|
| 452 |
+
"eval_accuracy": 0.821493227612126,
|
| 453 |
+
"eval_loss": 1.0429145097732544,
|
| 454 |
+
"eval_runtime": 229.2177,
|
| 455 |
+
"eval_samples_per_second": 89.147,
|
| 456 |
+
"eval_steps_per_second": 2.788,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 4.713914174252276,
|
| 461 |
+
"grad_norm": 3.1148979663848877,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.8871,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.713914174252276,
|
| 468 |
+
"eval_accuracy": 0.821055764186474,
|
| 469 |
+
"eval_loss": 1.0499247312545776,
|
| 470 |
+
"eval_runtime": 227.9461,
|
| 471 |
+
"eval_samples_per_second": 89.644,
|
| 472 |
+
"eval_steps_per_second": 2.803,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 4.876462938881664,
|
| 477 |
+
"grad_norm": 3.0099728107452393,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.8848,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 4.876462938881664,
|
| 484 |
+
"eval_accuracy": 0.8221091492282949,
|
| 485 |
+
"eval_loss": 1.0302263498306274,
|
| 486 |
+
"eval_runtime": 229.0246,
|
| 487 |
+
"eval_samples_per_second": 89.222,
|
| 488 |
+
"eval_steps_per_second": 2.79,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 5.039011703511053,
|
| 493 |
+
"grad_norm": 2.8249874114990234,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.8772,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 5.039011703511053,
|
| 500 |
+
"eval_accuracy": 0.8225020975532324,
|
| 501 |
+
"eval_loss": 1.0407395362854004,
|
| 502 |
+
"eval_runtime": 230.4007,
|
| 503 |
+
"eval_samples_per_second": 88.689,
|
| 504 |
+
"eval_steps_per_second": 2.773,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.201560468140442,
|
| 509 |
+
"grad_norm": 2.273552179336548,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.8715,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.201560468140442,
|
| 516 |
+
"eval_accuracy": 0.8238052823384368,
|
| 517 |
+
"eval_loss": 1.0325181484222412,
|
| 518 |
+
"eval_runtime": 230.5347,
|
| 519 |
+
"eval_samples_per_second": 88.637,
|
| 520 |
+
"eval_steps_per_second": 2.772,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 5.364109232769831,
|
| 525 |
+
"grad_norm": 2.884051561355591,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.8687,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 5.364109232769831,
|
| 532 |
+
"eval_accuracy": 0.8243404153748926,
|
| 533 |
+
"eval_loss": 1.0298871994018555,
|
| 534 |
+
"eval_runtime": 231.4091,
|
| 535 |
+
"eval_samples_per_second": 88.302,
|
| 536 |
+
"eval_steps_per_second": 2.761,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.526657997399219,
|
| 541 |
+
"grad_norm": 3.12709379196167,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.8674,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 5.526657997399219,
|
| 548 |
+
"eval_accuracy": 0.8245251637171149,
|
| 549 |
+
"eval_loss": 1.0233651399612427,
|
| 550 |
+
"eval_runtime": 225.4766,
|
| 551 |
+
"eval_samples_per_second": 90.626,
|
| 552 |
+
"eval_steps_per_second": 2.834,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 5.689206762028609,
|
| 557 |
+
"grad_norm": 2.7715749740600586,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.8597,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.689206762028609,
|
| 564 |
+
"eval_accuracy": 0.8253348922363006,
|
| 565 |
+
"eval_loss": 1.0265984535217285,
|
| 566 |
+
"eval_runtime": 226.4158,
|
| 567 |
+
"eval_samples_per_second": 90.25,
|
| 568 |
+
"eval_steps_per_second": 2.822,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 5.851755526657997,
|
| 573 |
+
"grad_norm": 2.630910873413086,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.8621,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 5.851755526657997,
|
| 580 |
+
"eval_accuracy": 0.8264286938861036,
|
| 581 |
+
"eval_loss": 1.025406002998352,
|
| 582 |
+
"eval_runtime": 225.265,
|
| 583 |
+
"eval_samples_per_second": 90.711,
|
| 584 |
+
"eval_steps_per_second": 2.837,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.014304291287386,
|
| 589 |
+
"grad_norm": 2.960498571395874,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.8505,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 6.014304291287386,
|
| 596 |
+
"eval_accuracy": 0.8266802737227764,
|
| 597 |
+
"eval_loss": 1.0281902551651,
|
| 598 |
+
"eval_runtime": 225.5813,
|
| 599 |
+
"eval_samples_per_second": 90.584,
|
| 600 |
+
"eval_steps_per_second": 2.833,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 6.176853055916775,
|
| 605 |
+
"grad_norm": 2.7250943183898926,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.8434,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 6.176853055916775,
|
| 612 |
+
"eval_accuracy": 0.8273285977403005,
|
| 613 |
+
"eval_loss": 1.019544005393982,
|
| 614 |
+
"eval_runtime": 225.1972,
|
| 615 |
+
"eval_samples_per_second": 90.738,
|
| 616 |
+
"eval_steps_per_second": 2.838,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 6.339401820546164,
|
| 621 |
+
"grad_norm": 2.45147967338562,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.8396,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.339401820546164,
|
| 628 |
+
"eval_accuracy": 0.8278276606616609,
|
| 629 |
+
"eval_loss": 1.016022801399231,
|
| 630 |
+
"eval_runtime": 225.1071,
|
| 631 |
+
"eval_samples_per_second": 90.775,
|
| 632 |
+
"eval_steps_per_second": 2.839,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.501950585175552,
|
| 637 |
+
"grad_norm": 3.0252089500427246,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.849,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.501950585175552,
|
| 644 |
+
"eval_accuracy": 0.8286753807254127,
|
| 645 |
+
"eval_loss": 0.9984493851661682,
|
| 646 |
+
"eval_runtime": 225.6293,
|
| 647 |
+
"eval_samples_per_second": 90.564,
|
| 648 |
+
"eval_steps_per_second": 2.832,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 6.664499349804942,
|
| 653 |
+
"grad_norm": 2.79874849319458,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.8344,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 6.664499349804942,
|
| 660 |
+
"eval_accuracy": 0.8295993429023929,
|
| 661 |
+
"eval_loss": 1.008083462715149,
|
| 662 |
+
"eval_runtime": 226.4394,
|
| 663 |
+
"eval_samples_per_second": 90.24,
|
| 664 |
+
"eval_steps_per_second": 2.822,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.82704811443433,
|
| 669 |
+
"grad_norm": 2.726980447769165,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.832,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.82704811443433,
|
| 676 |
+
"eval_accuracy": 0.8289772346442087,
|
| 677 |
+
"eval_loss": 1.0009534358978271,
|
| 678 |
+
"eval_runtime": 225.2727,
|
| 679 |
+
"eval_samples_per_second": 90.708,
|
| 680 |
+
"eval_steps_per_second": 2.837,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.989596879063719,
|
| 685 |
+
"grad_norm": 2.6177260875701904,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.8371,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 6.989596879063719,
|
| 692 |
+
"eval_accuracy": 0.828972862081629,
|
| 693 |
+
"eval_loss": 1.007731556892395,
|
| 694 |
+
"eval_runtime": 225.1985,
|
| 695 |
+
"eval_samples_per_second": 90.738,
|
| 696 |
+
"eval_steps_per_second": 2.837,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 7.152145643693108,
|
| 701 |
+
"grad_norm": 2.511810541152954,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.8221,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 7.152145643693108,
|
| 708 |
+
"eval_accuracy": 0.8296968824079836,
|
| 709 |
+
"eval_loss": 1.001420259475708,
|
| 710 |
+
"eval_runtime": 226.3788,
|
| 711 |
+
"eval_samples_per_second": 90.265,
|
| 712 |
+
"eval_steps_per_second": 2.823,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 7.314694408322497,
|
| 717 |
+
"grad_norm": 2.490196466445923,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.819,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 7.314694408322497,
|
| 724 |
+
"eval_accuracy": 0.8315370661967273,
|
| 725 |
+
"eval_loss": 1.0057367086410522,
|
| 726 |
+
"eval_runtime": 226.9835,
|
| 727 |
+
"eval_samples_per_second": 90.024,
|
| 728 |
+
"eval_steps_per_second": 2.815,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.477243172951885,
|
| 733 |
+
"grad_norm": 2.7740466594696045,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.8211,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 7.477243172951885,
|
| 740 |
+
"eval_accuracy": 0.8325044106908798,
|
| 741 |
+
"eval_loss": 0.9906172752380371,
|
| 742 |
+
"eval_runtime": 226.1902,
|
| 743 |
+
"eval_samples_per_second": 90.34,
|
| 744 |
+
"eval_steps_per_second": 2.825,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 7.639791937581275,
|
| 749 |
+
"grad_norm": 3.0789926052093506,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.818,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 7.639791937581275,
|
| 756 |
+
"eval_accuracy": 0.832777941957673,
|
| 757 |
+
"eval_loss": 0.9794951677322388,
|
| 758 |
+
"eval_runtime": 226.5187,
|
| 759 |
+
"eval_samples_per_second": 90.209,
|
| 760 |
+
"eval_steps_per_second": 2.821,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 7.802340702210663,
|
| 765 |
+
"grad_norm": 2.452514171600342,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.8128,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 7.802340702210663,
|
| 772 |
+
"eval_accuracy": 0.8325327815535476,
|
| 773 |
+
"eval_loss": 0.9824967384338379,
|
| 774 |
+
"eval_runtime": 225.9154,
|
| 775 |
+
"eval_samples_per_second": 90.45,
|
| 776 |
+
"eval_steps_per_second": 2.828,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 7.964889466840052,
|
| 781 |
+
"grad_norm": 2.486287832260132,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.8133,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 7.964889466840052,
|
| 788 |
+
"eval_accuracy": 0.8329719850198115,
|
| 789 |
+
"eval_loss": 0.9840763211250305,
|
| 790 |
+
"eval_runtime": 225.7676,
|
| 791 |
+
"eval_samples_per_second": 90.509,
|
| 792 |
+
"eval_steps_per_second": 2.83,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 8.12743823146944,
|
| 797 |
+
"grad_norm": 2.6238162517547607,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.7972,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 8.12743823146944,
|
| 804 |
+
"eval_accuracy": 0.8329068936503207,
|
| 805 |
+
"eval_loss": 0.9786662459373474,
|
| 806 |
+
"eval_runtime": 227.0649,
|
| 807 |
+
"eval_samples_per_second": 89.992,
|
| 808 |
+
"eval_steps_per_second": 2.814,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.289986996098829,
|
| 813 |
+
"grad_norm": 2.622385025024414,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.8012,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 8.289986996098829,
|
| 820 |
+
"eval_accuracy": 0.83353766380846,
|
| 821 |
+
"eval_loss": 0.9908860325813293,
|
| 822 |
+
"eval_runtime": 225.4511,
|
| 823 |
+
"eval_samples_per_second": 90.636,
|
| 824 |
+
"eval_steps_per_second": 2.834,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.45253576072822,
|
| 829 |
+
"grad_norm": 2.3659889698028564,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.8002,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 8.45253576072822,
|
| 836 |
+
"eval_accuracy": 0.834269520268432,
|
| 837 |
+
"eval_loss": 0.9886682629585266,
|
| 838 |
+
"eval_runtime": 226.2835,
|
| 839 |
+
"eval_samples_per_second": 90.303,
|
| 840 |
+
"eval_steps_per_second": 2.824,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 8.615084525357608,
|
| 845 |
+
"grad_norm": 2.6340997219085693,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.796,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 8.615084525357608,
|
| 852 |
+
"eval_accuracy": 0.83431812146113,
|
| 853 |
+
"eval_loss": 1.1027251482009888,
|
| 854 |
+
"eval_runtime": 224.7718,
|
| 855 |
+
"eval_samples_per_second": 90.91,
|
| 856 |
+
"eval_steps_per_second": 2.843,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 8.777633289986996,
|
| 861 |
+
"grad_norm": 2.351562976837158,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.8032,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 8.777633289986996,
|
| 868 |
+
"eval_accuracy": 0.83473517326524,
|
| 869 |
+
"eval_loss": 0.9701533317565918,
|
| 870 |
+
"eval_runtime": 224.7228,
|
| 871 |
+
"eval_samples_per_second": 90.93,
|
| 872 |
+
"eval_steps_per_second": 2.844,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 8.940182054616384,
|
| 877 |
+
"grad_norm": 3.179107189178467,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.7956,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 8.940182054616384,
|
| 884 |
+
"eval_accuracy": 0.8351845716261748,
|
| 885 |
+
"eval_loss": 0.978928804397583,
|
| 886 |
+
"eval_runtime": 226.3807,
|
| 887 |
+
"eval_samples_per_second": 90.264,
|
| 888 |
+
"eval_steps_per_second": 2.823,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 9.102730819245775,
|
| 893 |
+
"grad_norm": 2.7947020530700684,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.7897,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 9.102730819245775,
|
| 900 |
+
"eval_accuracy": 0.8353527515006416,
|
| 901 |
+
"eval_loss": 0.9703938961029053,
|
| 902 |
+
"eval_runtime": 227.4785,
|
| 903 |
+
"eval_samples_per_second": 89.828,
|
| 904 |
+
"eval_steps_per_second": 2.809,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 9.265279583875163,
|
| 909 |
+
"grad_norm": 2.4941184520721436,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.7822,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 9.265279583875163,
|
| 916 |
+
"eval_accuracy": 0.8354081248952335,
|
| 917 |
+
"eval_loss": 0.985792875289917,
|
| 918 |
+
"eval_runtime": 225.177,
|
| 919 |
+
"eval_samples_per_second": 90.746,
|
| 920 |
+
"eval_steps_per_second": 2.838,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 9.427828348504551,
|
| 925 |
+
"grad_norm": 2.623500108718872,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.7885,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 9.427828348504551,
|
| 932 |
+
"eval_accuracy": 0.8367873538892325,
|
| 933 |
+
"eval_loss": 0.9710731506347656,
|
| 934 |
+
"eval_runtime": 224.2999,
|
| 935 |
+
"eval_samples_per_second": 91.101,
|
| 936 |
+
"eval_steps_per_second": 2.849,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 9.59037711313394,
|
| 941 |
+
"grad_norm": 2.687804698944092,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.7853,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 9.59037711313394,
|
| 948 |
+
"eval_accuracy": 0.8370918540482961,
|
| 949 |
+
"eval_loss": 0.9488577246665955,
|
| 950 |
+
"eval_runtime": 225.5849,
|
| 951 |
+
"eval_samples_per_second": 90.582,
|
| 952 |
+
"eval_steps_per_second": 2.833,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 9.752925877763328,
|
| 957 |
+
"grad_norm": 2.415875196456909,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.7834,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 9.752925877763328,
|
| 964 |
+
"eval_accuracy": 0.8376393865857097,
|
| 965 |
+
"eval_loss": 0.9489056468009949,
|
| 966 |
+
"eval_runtime": 224.5743,
|
| 967 |
+
"eval_samples_per_second": 90.99,
|
| 968 |
+
"eval_steps_per_second": 2.845,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 9.915474642392718,
|
| 973 |
+
"grad_norm": 2.635620355606079,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.7795,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 9.915474642392718,
|
| 980 |
+
"eval_accuracy": 0.8379052346488646,
|
| 981 |
+
"eval_loss": 0.9600683450698853,
|
| 982 |
+
"eval_runtime": 227.5569,
|
| 983 |
+
"eval_samples_per_second": 89.797,
|
| 984 |
+
"eval_steps_per_second": 2.808,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 10.078023407022107,
|
| 989 |
+
"grad_norm": 2.2729740142822266,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.7784,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 10.078023407022107,
|
| 996 |
+
"eval_accuracy": 0.8387711597381718,
|
| 997 |
+
"eval_loss": 0.9676991701126099,
|
| 998 |
+
"eval_runtime": 227.0441,
|
| 999 |
+
"eval_samples_per_second": 90.0,
|
| 1000 |
+
"eval_steps_per_second": 2.814,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 10.240572171651495,
|
| 1005 |
+
"grad_norm": 2.665194511413574,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.7731,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 10.240572171651495,
|
| 1012 |
+
"eval_accuracy": 0.8384039380286287,
|
| 1013 |
+
"eval_loss": 0.9660561084747314,
|
| 1014 |
+
"eval_runtime": 225.7122,
|
| 1015 |
+
"eval_samples_per_second": 90.531,
|
| 1016 |
+
"eval_steps_per_second": 2.831,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 10.403120936280883,
|
| 1021 |
+
"grad_norm": 2.378326892852783,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.7732,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 10.403120936280883,
|
| 1028 |
+
"eval_accuracy": 0.8384703620413777,
|
| 1029 |
+
"eval_loss": 0.9522095322608948,
|
| 1030 |
+
"eval_runtime": 226.2237,
|
| 1031 |
+
"eval_samples_per_second": 90.327,
|
| 1032 |
+
"eval_steps_per_second": 2.825,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 10.565669700910274,
|
| 1037 |
+
"grad_norm": 2.3873043060302734,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.7675,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 10.565669700910274,
|
| 1044 |
+
"eval_accuracy": 0.8387303829039866,
|
| 1045 |
+
"eval_loss": 0.9506338238716125,
|
| 1046 |
+
"eval_runtime": 225.3705,
|
| 1047 |
+
"eval_samples_per_second": 90.668,
|
| 1048 |
+
"eval_steps_per_second": 2.835,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 10.728218465539662,
|
| 1053 |
+
"grad_norm": 2.908627986907959,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.7632,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 10.728218465539662,
|
| 1060 |
+
"eval_accuracy": 0.8393342612931991,
|
| 1061 |
+
"eval_loss": 0.9406708478927612,
|
| 1062 |
+
"eval_runtime": 226.5163,
|
| 1063 |
+
"eval_samples_per_second": 90.21,
|
| 1064 |
+
"eval_steps_per_second": 2.821,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 10.89076723016905,
|
| 1069 |
+
"grad_norm": 2.353163242340088,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.7652,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 10.89076723016905,
|
| 1076 |
+
"eval_accuracy": 0.8397924279467366,
|
| 1077 |
+
"eval_loss": 0.9450008869171143,
|
| 1078 |
+
"eval_runtime": 228.8699,
|
| 1079 |
+
"eval_samples_per_second": 89.282,
|
| 1080 |
+
"eval_steps_per_second": 2.792,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 11.053315994798439,
|
| 1085 |
+
"grad_norm": 2.5943007469177246,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.7569,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 11.053315994798439,
|
| 1092 |
+
"eval_accuracy": 0.8403287583785509,
|
| 1093 |
+
"eval_loss": 0.9399783611297607,
|
| 1094 |
+
"eval_runtime": 226.1121,
|
| 1095 |
+
"eval_samples_per_second": 90.371,
|
| 1096 |
+
"eval_steps_per_second": 2.826,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 11.215864759427829,
|
| 1101 |
+
"grad_norm": 2.693864107131958,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.7594,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 11.215864759427829,
|
| 1108 |
+
"eval_accuracy": 0.8401560936743129,
|
| 1109 |
+
"eval_loss": 0.9480236172676086,
|
| 1110 |
+
"eval_runtime": 225.4517,
|
| 1111 |
+
"eval_samples_per_second": 90.636,
|
| 1112 |
+
"eval_steps_per_second": 2.834,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 11.378413524057217,
|
| 1117 |
+
"grad_norm": 2.644212484359741,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.7549,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 11.378413524057217,
|
| 1124 |
+
"eval_accuracy": 0.8397807491674708,
|
| 1125 |
+
"eval_loss": 0.947754442691803,
|
| 1126 |
+
"eval_runtime": 224.7666,
|
| 1127 |
+
"eval_samples_per_second": 90.912,
|
| 1128 |
+
"eval_steps_per_second": 2.843,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 11.540962288686606,
|
| 1133 |
+
"grad_norm": 2.548055410385132,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.7508,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 11.540962288686606,
|
| 1140 |
+
"eval_accuracy": 0.8408923535971057,
|
| 1141 |
+
"eval_loss": 0.9363004565238953,
|
| 1142 |
+
"eval_runtime": 226.2948,
|
| 1143 |
+
"eval_samples_per_second": 90.298,
|
| 1144 |
+
"eval_steps_per_second": 2.824,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 11.703511053315994,
|
| 1149 |
+
"grad_norm": 2.2983460426330566,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.7498,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 11.703511053315994,
|
| 1156 |
+
"eval_accuracy": 0.8415923695179955,
|
| 1157 |
+
"eval_loss": 0.9559971690177917,
|
| 1158 |
+
"eval_runtime": 225.5228,
|
| 1159 |
+
"eval_samples_per_second": 90.607,
|
| 1160 |
+
"eval_steps_per_second": 2.833,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 11.866059817945384,
|
| 1165 |
+
"grad_norm": 2.585132122039795,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.7529,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 11.866059817945384,
|
| 1172 |
+
"eval_accuracy": 0.8414500365516611,
|
| 1173 |
+
"eval_loss": 0.9397149085998535,
|
| 1174 |
+
"eval_runtime": 225.5184,
|
| 1175 |
+
"eval_samples_per_second": 90.609,
|
| 1176 |
+
"eval_steps_per_second": 2.833,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 12.028608582574773,
|
| 1181 |
+
"grad_norm": 2.7008893489837646,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.7499,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 12.028608582574773,
|
| 1188 |
+
"eval_accuracy": 0.8429048722887895,
|
| 1189 |
+
"eval_loss": 0.9322752952575684,
|
| 1190 |
+
"eval_runtime": 226.0814,
|
| 1191 |
+
"eval_samples_per_second": 90.383,
|
| 1192 |
+
"eval_steps_per_second": 2.826,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 12.191157347204161,
|
| 1197 |
+
"grad_norm": 2.234344959259033,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.7516,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 12.191157347204161,
|
| 1204 |
+
"eval_accuracy": 0.8423553938504648,
|
| 1205 |
+
"eval_loss": 0.9274143576622009,
|
| 1206 |
+
"eval_runtime": 224.6544,
|
| 1207 |
+
"eval_samples_per_second": 90.957,
|
| 1208 |
+
"eval_steps_per_second": 2.844,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 12.35370611183355,
|
| 1213 |
+
"grad_norm": 2.38130259513855,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.742,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 12.35370611183355,
|
| 1220 |
+
"eval_accuracy": 0.8434239606449652,
|
| 1221 |
+
"eval_loss": 0.9298020005226135,
|
| 1222 |
+
"eval_runtime": 224.8881,
|
| 1223 |
+
"eval_samples_per_second": 90.863,
|
| 1224 |
+
"eval_steps_per_second": 2.841,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 12.51625487646294,
|
| 1229 |
+
"grad_norm": 2.4753472805023193,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.7349,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 12.51625487646294,
|
| 1236 |
+
"eval_accuracy": 0.8435835943839733,
|
| 1237 |
+
"eval_loss": 0.9265833497047424,
|
| 1238 |
+
"eval_runtime": 225.4856,
|
| 1239 |
+
"eval_samples_per_second": 90.622,
|
| 1240 |
+
"eval_steps_per_second": 2.834,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 12.678803641092328,
|
| 1245 |
+
"grad_norm": 2.6576738357543945,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.7408,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 12.678803641092328,
|
| 1252 |
+
"eval_accuracy": 0.8433502485604528,
|
| 1253 |
+
"eval_loss": 0.9311428070068359,
|
| 1254 |
+
"eval_runtime": 226.2881,
|
| 1255 |
+
"eval_samples_per_second": 90.301,
|
| 1256 |
+
"eval_steps_per_second": 2.824,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 12.841352405721716,
|
| 1261 |
+
"grad_norm": 2.318998336791992,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.742,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 12.841352405721716,
|
| 1268 |
+
"eval_accuracy": 0.8439347116351804,
|
| 1269 |
+
"eval_loss": 0.9278241991996765,
|
| 1270 |
+
"eval_runtime": 226.17,
|
| 1271 |
+
"eval_samples_per_second": 90.348,
|
| 1272 |
+
"eval_steps_per_second": 2.825,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 13.003901170351105,
|
| 1277 |
+
"grad_norm": 2.5888776779174805,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.735,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 13.003901170351105,
|
| 1284 |
+
"eval_accuracy": 0.8435309371055906,
|
| 1285 |
+
"eval_loss": 0.9364249110221863,
|
| 1286 |
+
"eval_runtime": 226.5769,
|
| 1287 |
+
"eval_samples_per_second": 90.186,
|
| 1288 |
+
"eval_steps_per_second": 2.82,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 13.166449934980495,
|
| 1293 |
+
"grad_norm": 2.5382964611053467,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.7317,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 13.166449934980495,
|
| 1300 |
+
"eval_accuracy": 0.8435685143232774,
|
| 1301 |
+
"eval_loss": 0.9289753437042236,
|
| 1302 |
+
"eval_runtime": 226.0444,
|
| 1303 |
+
"eval_samples_per_second": 90.398,
|
| 1304 |
+
"eval_steps_per_second": 2.827,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 13.328998699609883,
|
| 1309 |
+
"grad_norm": 2.3561949729919434,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.7287,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 13.328998699609883,
|
| 1316 |
+
"eval_accuracy": 0.8441468378237561,
|
| 1317 |
+
"eval_loss": 0.9176314473152161,
|
| 1318 |
+
"eval_runtime": 225.4845,
|
| 1319 |
+
"eval_samples_per_second": 90.623,
|
| 1320 |
+
"eval_steps_per_second": 2.834,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 13.491547464239272,
|
| 1325 |
+
"grad_norm": 2.4950144290924072,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.7236,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 13.491547464239272,
|
| 1332 |
+
"eval_accuracy": 0.8440490876509515,
|
| 1333 |
+
"eval_loss": 0.9126904010772705,
|
| 1334 |
+
"eval_runtime": 225.1525,
|
| 1335 |
+
"eval_samples_per_second": 90.756,
|
| 1336 |
+
"eval_steps_per_second": 2.838,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 13.65409622886866,
|
| 1341 |
+
"grad_norm": 2.5094358921051025,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.7271,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 13.65409622886866,
|
| 1348 |
+
"eval_accuracy": 0.844420508518643,
|
| 1349 |
+
"eval_loss": 0.9222070574760437,
|
| 1350 |
+
"eval_runtime": 226.8881,
|
| 1351 |
+
"eval_samples_per_second": 90.062,
|
| 1352 |
+
"eval_steps_per_second": 2.816,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 13.81664499349805,
|
| 1357 |
+
"grad_norm": 2.5358660221099854,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.7247,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 13.81664499349805,
|
| 1364 |
+
"eval_accuracy": 0.8447406385701589,
|
| 1365 |
+
"eval_loss": 0.911141037940979,
|
| 1366 |
+
"eval_runtime": 226.4876,
|
| 1367 |
+
"eval_samples_per_second": 90.221,
|
| 1368 |
+
"eval_steps_per_second": 2.821,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 13.979193758127439,
|
| 1373 |
+
"grad_norm": 2.449117422103882,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.724,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 13.979193758127439,
|
| 1380 |
+
"eval_accuracy": 0.8455180692523965,
|
| 1381 |
+
"eval_loss": 0.9233959317207336,
|
| 1382 |
+
"eval_runtime": 226.6454,
|
| 1383 |
+
"eval_samples_per_second": 90.158,
|
| 1384 |
+
"eval_steps_per_second": 2.819,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 14.141742522756827,
|
| 1389 |
+
"grad_norm": 2.9299075603485107,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.717,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 14.141742522756827,
|
| 1396 |
+
"eval_accuracy": 0.8457589583161474,
|
| 1397 |
+
"eval_loss": 0.9212149381637573,
|
| 1398 |
+
"eval_runtime": 226.3542,
|
| 1399 |
+
"eval_samples_per_second": 90.274,
|
| 1400 |
+
"eval_steps_per_second": 2.823,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 14.304291287386215,
|
| 1405 |
+
"grad_norm": 2.705927610397339,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.7209,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 14.304291287386215,
|
| 1412 |
+
"eval_accuracy": 0.8461540554265815,
|
| 1413 |
+
"eval_loss": 0.9096006155014038,
|
| 1414 |
+
"eval_runtime": 225.3071,
|
| 1415 |
+
"eval_samples_per_second": 90.694,
|
| 1416 |
+
"eval_steps_per_second": 2.836,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 14.466840052015606,
|
| 1421 |
+
"grad_norm": 2.0237460136413574,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.7143,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 14.466840052015606,
|
| 1428 |
+
"eval_accuracy": 0.8467122032637585,
|
| 1429 |
+
"eval_loss": 0.9058763980865479,
|
| 1430 |
+
"eval_runtime": 224.7066,
|
| 1431 |
+
"eval_samples_per_second": 90.936,
|
| 1432 |
+
"eval_steps_per_second": 2.844,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 14.629388816644994,
|
| 1437 |
+
"grad_norm": 2.6412603855133057,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.7097,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 14.629388816644994,
|
| 1444 |
+
"eval_accuracy": 0.8469896352526538,
|
| 1445 |
+
"eval_loss": 0.9109433889389038,
|
| 1446 |
+
"eval_runtime": 226.4374,
|
| 1447 |
+
"eval_samples_per_second": 90.241,
|
| 1448 |
+
"eval_steps_per_second": 2.822,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 14.791937581274382,
|
| 1453 |
+
"grad_norm": 2.0750892162323,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.7167,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 14.791937581274382,
|
| 1460 |
+
"eval_accuracy": 0.8459233510762237,
|
| 1461 |
+
"eval_loss": 0.9634975790977478,
|
| 1462 |
+
"eval_runtime": 224.623,
|
| 1463 |
+
"eval_samples_per_second": 90.97,
|
| 1464 |
+
"eval_steps_per_second": 2.845,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 14.95448634590377,
|
| 1469 |
+
"grad_norm": 3.2299835681915283,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.7175,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 14.95448634590377,
|
| 1476 |
+
"eval_accuracy": 0.8472973119877086,
|
| 1477 |
+
"eval_loss": 0.9039083123207092,
|
| 1478 |
+
"eval_runtime": 226.4661,
|
| 1479 |
+
"eval_samples_per_second": 90.23,
|
| 1480 |
+
"eval_steps_per_second": 2.822,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 15.117035110533159,
|
| 1485 |
+
"grad_norm": 2.562220573425293,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.7032,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 15.117035110533159,
|
| 1492 |
+
"eval_accuracy": 0.8473520786144324,
|
| 1493 |
+
"eval_loss": 0.911507785320282,
|
| 1494 |
+
"eval_runtime": 224.8892,
|
| 1495 |
+
"eval_samples_per_second": 90.863,
|
| 1496 |
+
"eval_steps_per_second": 2.841,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 15.27958387516255,
|
| 1501 |
+
"grad_norm": 2.8594844341278076,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.7082,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 15.27958387516255,
|
| 1508 |
+
"eval_accuracy": 0.8476350142048673,
|
| 1509 |
+
"eval_loss": 0.909827470779419,
|
| 1510 |
+
"eval_runtime": 225.7006,
|
| 1511 |
+
"eval_samples_per_second": 90.536,
|
| 1512 |
+
"eval_steps_per_second": 2.831,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 15.442132639791938,
|
| 1517 |
+
"grad_norm": 2.468425989151001,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.7041,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 15.442132639791938,
|
| 1524 |
+
"eval_accuracy": 0.8478791054352419,
|
| 1525 |
+
"eval_loss": 0.9049926996231079,
|
| 1526 |
+
"eval_runtime": 224.8771,
|
| 1527 |
+
"eval_samples_per_second": 90.867,
|
| 1528 |
+
"eval_steps_per_second": 2.842,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 15.604681404421326,
|
| 1533 |
+
"grad_norm": 2.389726400375366,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.7031,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 15.604681404421326,
|
| 1540 |
+
"eval_accuracy": 0.8481061886894397,
|
| 1541 |
+
"eval_loss": 0.890941858291626,
|
| 1542 |
+
"eval_runtime": 226.989,
|
| 1543 |
+
"eval_samples_per_second": 90.022,
|
| 1544 |
+
"eval_steps_per_second": 2.815,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 15.767230169050714,
|
| 1549 |
+
"grad_norm": 2.2124507427215576,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.6986,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 15.767230169050714,
|
| 1556 |
+
"eval_accuracy": 0.8491361761433549,
|
| 1557 |
+
"eval_loss": 0.9069699048995972,
|
| 1558 |
+
"eval_runtime": 224.9486,
|
| 1559 |
+
"eval_samples_per_second": 90.839,
|
| 1560 |
+
"eval_steps_per_second": 2.841,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 15.929778933680105,
|
| 1565 |
+
"grad_norm": 2.2906134128570557,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.707,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 15.929778933680105,
|
| 1572 |
+
"eval_accuracy": 0.8479968642310016,
|
| 1573 |
+
"eval_loss": 0.9036478996276855,
|
| 1574 |
+
"eval_runtime": 225.9102,
|
| 1575 |
+
"eval_samples_per_second": 90.452,
|
| 1576 |
+
"eval_steps_per_second": 2.829,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 16.092327698309493,
|
| 1581 |
+
"grad_norm": 2.6677193641662598,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.6948,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 16.092327698309493,
|
| 1588 |
+
"eval_accuracy": 0.8493272111938716,
|
| 1589 |
+
"eval_loss": 0.9056188464164734,
|
| 1590 |
+
"eval_runtime": 225.0003,
|
| 1591 |
+
"eval_samples_per_second": 90.818,
|
| 1592 |
+
"eval_steps_per_second": 2.84,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 16.25487646293888,
|
| 1597 |
+
"grad_norm": 2.4262144565582275,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.6967,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 16.25487646293888,
|
| 1604 |
+
"eval_accuracy": 0.8485872336104957,
|
| 1605 |
+
"eval_loss": 0.896799623966217,
|
| 1606 |
+
"eval_runtime": 224.1595,
|
| 1607 |
+
"eval_samples_per_second": 91.158,
|
| 1608 |
+
"eval_steps_per_second": 2.851,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 16.41742522756827,
|
| 1613 |
+
"grad_norm": 2.316122055053711,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.6977,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 16.41742522756827,
|
| 1620 |
+
"eval_accuracy": 0.8495556837644054,
|
| 1621 |
+
"eval_loss": 0.9065157175064087,
|
| 1622 |
+
"eval_runtime": 225.8926,
|
| 1623 |
+
"eval_samples_per_second": 90.459,
|
| 1624 |
+
"eval_steps_per_second": 2.829,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 16.579973992197658,
|
| 1629 |
+
"grad_norm": 2.6691653728485107,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.6961,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 16.579973992197658,
|
| 1636 |
+
"eval_accuracy": 0.8500652800696321,
|
| 1637 |
+
"eval_loss": 0.896096408367157,
|
| 1638 |
+
"eval_runtime": 232.2659,
|
| 1639 |
+
"eval_samples_per_second": 87.977,
|
| 1640 |
+
"eval_steps_per_second": 2.751,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 16.742522756827046,
|
| 1645 |
+
"grad_norm": 2.475905656814575,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.6916,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 16.742522756827046,
|
| 1652 |
+
"eval_accuracy": 0.8497042599251818,
|
| 1653 |
+
"eval_loss": 0.8976150155067444,
|
| 1654 |
+
"eval_runtime": 226.0766,
|
| 1655 |
+
"eval_samples_per_second": 90.385,
|
| 1656 |
+
"eval_steps_per_second": 2.826,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 16.90507152145644,
|
| 1661 |
+
"grad_norm": 2.786370277404785,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.6914,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 16.90507152145644,
|
| 1668 |
+
"eval_accuracy": 0.8496545695743757,
|
| 1669 |
+
"eval_loss": 0.8938505053520203,
|
| 1670 |
+
"eval_runtime": 224.6085,
|
| 1671 |
+
"eval_samples_per_second": 90.976,
|
| 1672 |
+
"eval_steps_per_second": 2.845,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 17.067620286085827,
|
| 1677 |
+
"grad_norm": 2.8185195922851562,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.6843,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 17.067620286085827,
|
| 1684 |
+
"eval_accuracy": 0.8506024205488569,
|
| 1685 |
+
"eval_loss": 0.902123749256134,
|
| 1686 |
+
"eval_runtime": 225.1853,
|
| 1687 |
+
"eval_samples_per_second": 90.743,
|
| 1688 |
+
"eval_steps_per_second": 2.838,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 17.230169050715215,
|
| 1693 |
+
"grad_norm": 2.287614107131958,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.6886,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 17.230169050715215,
|
| 1700 |
+
"eval_accuracy": 0.8509516099496137,
|
| 1701 |
+
"eval_loss": 0.8869491815567017,
|
| 1702 |
+
"eval_runtime": 225.8223,
|
| 1703 |
+
"eval_samples_per_second": 90.487,
|
| 1704 |
+
"eval_steps_per_second": 2.83,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 17.392717815344604,
|
| 1709 |
+
"grad_norm": 2.545166492462158,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.6895,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 17.392717815344604,
|
| 1716 |
+
"eval_accuracy": 0.8509741315809264,
|
| 1717 |
+
"eval_loss": 0.88730388879776,
|
| 1718 |
+
"eval_runtime": 225.2327,
|
| 1719 |
+
"eval_samples_per_second": 90.724,
|
| 1720 |
+
"eval_steps_per_second": 2.837,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 17.555266579973992,
|
| 1725 |
+
"grad_norm": 2.133690357208252,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.684,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 17.555266579973992,
|
| 1732 |
+
"eval_accuracy": 0.8506065339944305,
|
| 1733 |
+
"eval_loss": 0.8796634674072266,
|
| 1734 |
+
"eval_runtime": 225.9145,
|
| 1735 |
+
"eval_samples_per_second": 90.45,
|
| 1736 |
+
"eval_steps_per_second": 2.829,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 17.71781534460338,
|
| 1741 |
+
"grad_norm": 2.368135690689087,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.6814,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 17.71781534460338,
|
| 1748 |
+
"eval_accuracy": 0.8518810256708096,
|
| 1749 |
+
"eval_loss": 0.8883038759231567,
|
| 1750 |
+
"eval_runtime": 226.0516,
|
| 1751 |
+
"eval_samples_per_second": 90.395,
|
| 1752 |
+
"eval_steps_per_second": 2.827,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 17.88036410923277,
|
| 1757 |
+
"grad_norm": 2.93131685256958,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.6814,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 17.88036410923277,
|
| 1764 |
+
"eval_accuracy": 0.8515069138621577,
|
| 1765 |
+
"eval_loss": 0.8804932832717896,
|
| 1766 |
+
"eval_runtime": 224.3083,
|
| 1767 |
+
"eval_samples_per_second": 91.098,
|
| 1768 |
+
"eval_steps_per_second": 2.849,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 18.042912873862157,
|
| 1773 |
+
"grad_norm": 2.6115188598632812,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.6825,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 18.042912873862157,
|
| 1780 |
+
"eval_accuracy": 0.8524721846400168,
|
| 1781 |
+
"eval_loss": 0.8792157173156738,
|
| 1782 |
+
"eval_runtime": 225.8719,
|
| 1783 |
+
"eval_samples_per_second": 90.467,
|
| 1784 |
+
"eval_steps_per_second": 2.829,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 18.20546163849155,
|
| 1789 |
+
"grad_norm": 3.650696039199829,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.6754,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 18.20546163849155,
|
| 1796 |
+
"eval_accuracy": 0.8515890870808295,
|
| 1797 |
+
"eval_loss": 0.8762602210044861,
|
| 1798 |
+
"eval_runtime": 225.4458,
|
| 1799 |
+
"eval_samples_per_second": 90.638,
|
| 1800 |
+
"eval_steps_per_second": 2.834,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 18.368010403120937,
|
| 1805 |
+
"grad_norm": 2.48811936378479,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.6694,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 18.368010403120937,
|
| 1812 |
+
"eval_accuracy": 0.852083188035357,
|
| 1813 |
+
"eval_loss": 0.8903321623802185,
|
| 1814 |
+
"eval_runtime": 231.1517,
|
| 1815 |
+
"eval_samples_per_second": 88.401,
|
| 1816 |
+
"eval_steps_per_second": 2.764,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 18.530559167750326,
|
| 1821 |
+
"grad_norm": 2.664206027984619,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.6711,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 18.530559167750326,
|
| 1828 |
+
"eval_accuracy": 0.8524635038150364,
|
| 1829 |
+
"eval_loss": 0.8883496522903442,
|
| 1830 |
+
"eval_runtime": 224.7773,
|
| 1831 |
+
"eval_samples_per_second": 90.908,
|
| 1832 |
+
"eval_steps_per_second": 2.843,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 18.693107932379714,
|
| 1837 |
+
"grad_norm": 2.5020461082458496,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.6701,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 18.693107932379714,
|
| 1844 |
+
"eval_accuracy": 0.8527774358008777,
|
| 1845 |
+
"eval_loss": 0.8737440705299377,
|
| 1846 |
+
"eval_runtime": 224.3031,
|
| 1847 |
+
"eval_samples_per_second": 91.1,
|
| 1848 |
+
"eval_steps_per_second": 2.849,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 18.855656697009103,
|
| 1853 |
+
"grad_norm": 2.341423273086548,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.6722,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 18.855656697009103,
|
| 1860 |
+
"eval_accuracy": 0.8529963243635703,
|
| 1861 |
+
"eval_loss": 0.8735175132751465,
|
| 1862 |
+
"eval_runtime": 225.4167,
|
| 1863 |
+
"eval_samples_per_second": 90.65,
|
| 1864 |
+
"eval_steps_per_second": 2.835,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 19.01820546163849,
|
| 1869 |
+
"grad_norm": 2.4827890396118164,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.6723,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 19.01820546163849,
|
| 1876 |
+
"eval_accuracy": 0.853864474597493,
|
| 1877 |
+
"eval_loss": 0.8714447021484375,
|
| 1878 |
+
"eval_runtime": 225.7934,
|
| 1879 |
+
"eval_samples_per_second": 90.499,
|
| 1880 |
+
"eval_steps_per_second": 2.83,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 19.18075422626788,
|
| 1885 |
+
"grad_norm": 2.6257293224334717,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.6623,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 19.18075422626788,
|
| 1892 |
+
"eval_accuracy": 0.8535918441752076,
|
| 1893 |
+
"eval_loss": 0.8707379698753357,
|
| 1894 |
+
"eval_runtime": 233.0012,
|
| 1895 |
+
"eval_samples_per_second": 87.699,
|
| 1896 |
+
"eval_steps_per_second": 2.742,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 19.343302990897268,
|
| 1901 |
+
"grad_norm": 2.2881245613098145,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.6629,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 19.343302990897268,
|
| 1908 |
+
"eval_accuracy": 0.8536284370228246,
|
| 1909 |
+
"eval_loss": 0.8575323820114136,
|
| 1910 |
+
"eval_runtime": 225.6251,
|
| 1911 |
+
"eval_samples_per_second": 90.566,
|
| 1912 |
+
"eval_steps_per_second": 2.832,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 19.505851755526656,
|
| 1917 |
+
"grad_norm": 2.485198497772217,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.6629,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 19.505851755526656,
|
| 1924 |
+
"eval_accuracy": 0.8541063694677179,
|
| 1925 |
+
"eval_loss": 0.8660362958908081,
|
| 1926 |
+
"eval_runtime": 229.1829,
|
| 1927 |
+
"eval_samples_per_second": 89.16,
|
| 1928 |
+
"eval_steps_per_second": 2.788,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 19.668400520156048,
|
| 1933 |
+
"grad_norm": 2.365445375442505,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.6636,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 19.668400520156048,
|
| 1940 |
+
"eval_accuracy": 0.8544101305781451,
|
| 1941 |
+
"eval_loss": 0.8751095533370972,
|
| 1942 |
+
"eval_runtime": 226.4875,
|
| 1943 |
+
"eval_samples_per_second": 90.221,
|
| 1944 |
+
"eval_steps_per_second": 2.821,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 19.830949284785437,
|
| 1949 |
+
"grad_norm": 2.7341248989105225,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.6668,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 19.830949284785437,
|
| 1956 |
+
"eval_accuracy": 0.8538102982406334,
|
| 1957 |
+
"eval_loss": 0.8749545812606812,
|
| 1958 |
+
"eval_runtime": 225.3871,
|
| 1959 |
+
"eval_samples_per_second": 90.662,
|
| 1960 |
+
"eval_steps_per_second": 2.835,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 19.993498049414825,
|
| 1965 |
+
"grad_norm": 2.200584888458252,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.6658,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 19.993498049414825,
|
| 1972 |
+
"eval_accuracy": 0.8542889960446646,
|
| 1973 |
+
"eval_loss": 0.8671930432319641,
|
| 1974 |
+
"eval_runtime": 225.4174,
|
| 1975 |
+
"eval_samples_per_second": 90.65,
|
| 1976 |
+
"eval_steps_per_second": 2.835,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 20.156046814044213,
|
| 1981 |
+
"grad_norm": 2.7894246578216553,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.6616,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 20.156046814044213,
|
| 1988 |
+
"eval_accuracy": 0.8548723750905142,
|
| 1989 |
+
"eval_loss": 0.8662538528442383,
|
| 1990 |
+
"eval_runtime": 224.781,
|
| 1991 |
+
"eval_samples_per_second": 90.906,
|
| 1992 |
+
"eval_steps_per_second": 2.843,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 20.3185955786736,
|
| 1997 |
+
"grad_norm": 2.3690404891967773,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.6541,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 20.3185955786736,
|
| 2004 |
+
"eval_accuracy": 0.8550845489091575,
|
| 2005 |
+
"eval_loss": 0.865816593170166,
|
| 2006 |
+
"eval_runtime": 228.0119,
|
| 2007 |
+
"eval_samples_per_second": 89.618,
|
| 2008 |
+
"eval_steps_per_second": 2.802,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 20.48114434330299,
|
| 2013 |
+
"grad_norm": 2.6345200538635254,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.6561,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 20.48114434330299,
|
| 2020 |
+
"eval_accuracy": 0.8558111308868751,
|
| 2021 |
+
"eval_loss": 0.8506866097450256,
|
| 2022 |
+
"eval_runtime": 225.0923,
|
| 2023 |
+
"eval_samples_per_second": 90.781,
|
| 2024 |
+
"eval_steps_per_second": 2.839,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 20.64369310793238,
|
| 2029 |
+
"grad_norm": 2.1237897872924805,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.6578,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 20.64369310793238,
|
| 2036 |
+
"eval_accuracy": 0.8554877275609093,
|
| 2037 |
+
"eval_loss": 0.8685211539268494,
|
| 2038 |
+
"eval_runtime": 225.3889,
|
| 2039 |
+
"eval_samples_per_second": 90.661,
|
| 2040 |
+
"eval_steps_per_second": 2.835,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 20.806241872561767,
|
| 2045 |
+
"grad_norm": 2.37259578704834,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.657,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 20.806241872561767,
|
| 2052 |
+
"eval_accuracy": 0.8553885784782709,
|
| 2053 |
+
"eval_loss": 0.8665293455123901,
|
| 2054 |
+
"eval_runtime": 225.5872,
|
| 2055 |
+
"eval_samples_per_second": 90.581,
|
| 2056 |
+
"eval_steps_per_second": 2.833,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 20.96879063719116,
|
| 2061 |
+
"grad_norm": 2.595446825027466,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.6495,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 20.96879063719116,
|
| 2068 |
+
"eval_accuracy": 0.8558350742146078,
|
| 2069 |
+
"eval_loss": 0.8600612878799438,
|
| 2070 |
+
"eval_runtime": 225.3383,
|
| 2071 |
+
"eval_samples_per_second": 90.681,
|
| 2072 |
+
"eval_steps_per_second": 2.836,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 21.131339401820547,
|
| 2077 |
+
"grad_norm": 2.689633846282959,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.6484,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 21.131339401820547,
|
| 2084 |
+
"eval_accuracy": 0.8563225017454439,
|
| 2085 |
+
"eval_loss": 0.8686124086380005,
|
| 2086 |
+
"eval_runtime": 225.5265,
|
| 2087 |
+
"eval_samples_per_second": 90.606,
|
| 2088 |
+
"eval_steps_per_second": 2.833,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 21.293888166449936,
|
| 2093 |
+
"grad_norm": 2.4909982681274414,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.6429,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 21.293888166449936,
|
| 2100 |
+
"eval_accuracy": 0.8557131214639457,
|
| 2101 |
+
"eval_loss": 0.86305832862854,
|
| 2102 |
+
"eval_runtime": 225.2213,
|
| 2103 |
+
"eval_samples_per_second": 90.729,
|
| 2104 |
+
"eval_steps_per_second": 2.837,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 21.456436931079324,
|
| 2109 |
+
"grad_norm": 2.7555160522460938,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.6523,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 21.456436931079324,
|
| 2116 |
+
"eval_accuracy": 0.8565990504258199,
|
| 2117 |
+
"eval_loss": 0.8672294616699219,
|
| 2118 |
+
"eval_runtime": 225.3745,
|
| 2119 |
+
"eval_samples_per_second": 90.667,
|
| 2120 |
+
"eval_steps_per_second": 2.835,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 21.618985695708712,
|
| 2125 |
+
"grad_norm": 2.4315414428710938,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.6384,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 21.618985695708712,
|
| 2132 |
+
"eval_accuracy": 0.8567521705315332,
|
| 2133 |
+
"eval_loss": 0.8545786738395691,
|
| 2134 |
+
"eval_runtime": 226.5184,
|
| 2135 |
+
"eval_samples_per_second": 90.209,
|
| 2136 |
+
"eval_steps_per_second": 2.821,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 21.7815344603381,
|
| 2141 |
+
"grad_norm": 2.373359441757202,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.6512,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 21.7815344603381,
|
| 2148 |
+
"eval_accuracy": 0.8568411511443462,
|
| 2149 |
+
"eval_loss": 0.8615005612373352,
|
| 2150 |
+
"eval_runtime": 226.7483,
|
| 2151 |
+
"eval_samples_per_second": 90.118,
|
| 2152 |
+
"eval_steps_per_second": 2.818,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 21.94408322496749,
|
| 2157 |
+
"grad_norm": 2.5687010288238525,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.6452,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 21.94408322496749,
|
| 2164 |
+
"eval_accuracy": 0.856976936965993,
|
| 2165 |
+
"eval_loss": 0.8514290452003479,
|
| 2166 |
+
"eval_runtime": 224.8277,
|
| 2167 |
+
"eval_samples_per_second": 90.887,
|
| 2168 |
+
"eval_steps_per_second": 2.842,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 22.106631989596877,
|
| 2173 |
+
"grad_norm": 2.3000574111938477,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.6388,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 22.106631989596877,
|
| 2180 |
+
"eval_accuracy": 0.8576112543945463,
|
| 2181 |
+
"eval_loss": 0.8523911833763123,
|
| 2182 |
+
"eval_runtime": 225.5646,
|
| 2183 |
+
"eval_samples_per_second": 90.59,
|
| 2184 |
+
"eval_steps_per_second": 2.833,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 22.26918075422627,
|
| 2189 |
+
"grad_norm": 2.5039753913879395,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.6382,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 22.26918075422627,
|
| 2196 |
+
"eval_accuracy": 0.8567441321838508,
|
| 2197 |
+
"eval_loss": 0.8560938835144043,
|
| 2198 |
+
"eval_runtime": 226.9717,
|
| 2199 |
+
"eval_samples_per_second": 90.029,
|
| 2200 |
+
"eval_steps_per_second": 2.815,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 22.431729518855658,
|
| 2205 |
+
"grad_norm": 2.6623408794403076,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.6408,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 22.431729518855658,
|
| 2212 |
+
"eval_accuracy": 0.8576559655770156,
|
| 2213 |
+
"eval_loss": 0.854763925075531,
|
| 2214 |
+
"eval_runtime": 226.0318,
|
| 2215 |
+
"eval_samples_per_second": 90.403,
|
| 2216 |
+
"eval_steps_per_second": 2.827,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 22.594278283485046,
|
| 2221 |
+
"grad_norm": 2.3223702907562256,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.6433,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 22.594278283485046,
|
| 2228 |
+
"eval_accuracy": 0.8581642158372993,
|
| 2229 |
+
"eval_loss": 0.8585615754127502,
|
| 2230 |
+
"eval_runtime": 225.6463,
|
| 2231 |
+
"eval_samples_per_second": 90.558,
|
| 2232 |
+
"eval_steps_per_second": 2.832,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 22.756827048114435,
|
| 2237 |
+
"grad_norm": 2.545360803604126,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.6371,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 22.756827048114435,
|
| 2244 |
+
"eval_accuracy": 0.8583454058165224,
|
| 2245 |
+
"eval_loss": 0.8482581973075867,
|
| 2246 |
+
"eval_runtime": 225.5921,
|
| 2247 |
+
"eval_samples_per_second": 90.579,
|
| 2248 |
+
"eval_steps_per_second": 2.833,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 22.919375812743823,
|
| 2253 |
+
"grad_norm": 2.18627667427063,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.6331,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 22.919375812743823,
|
| 2260 |
+
"eval_accuracy": 0.857949745807387,
|
| 2261 |
+
"eval_loss": 0.8507857918739319,
|
| 2262 |
+
"eval_runtime": 226.8073,
|
| 2263 |
+
"eval_samples_per_second": 90.094,
|
| 2264 |
+
"eval_steps_per_second": 2.817,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 23.08192457737321,
|
| 2269 |
+
"grad_norm": 2.8513870239257812,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.6393,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 23.08192457737321,
|
| 2276 |
+
"eval_accuracy": 0.8573411368207433,
|
| 2277 |
+
"eval_loss": 0.8502649664878845,
|
| 2278 |
+
"eval_runtime": 227.8606,
|
| 2279 |
+
"eval_samples_per_second": 89.678,
|
| 2280 |
+
"eval_steps_per_second": 2.804,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 23.2444733420026,
|
| 2285 |
+
"grad_norm": 2.961089611053467,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.6269,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 23.2444733420026,
|
| 2292 |
+
"eval_accuracy": 0.8582647966293633,
|
| 2293 |
+
"eval_loss": 0.8488523364067078,
|
| 2294 |
+
"eval_runtime": 226.4431,
|
| 2295 |
+
"eval_samples_per_second": 90.239,
|
| 2296 |
+
"eval_steps_per_second": 2.822,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 23.407022106631988,
|
| 2301 |
+
"grad_norm": 2.448005437850952,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.6284,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 23.407022106631988,
|
| 2308 |
+
"eval_accuracy": 0.8588356517259145,
|
| 2309 |
+
"eval_loss": 0.8428735136985779,
|
| 2310 |
+
"eval_runtime": 226.3587,
|
| 2311 |
+
"eval_samples_per_second": 90.273,
|
| 2312 |
+
"eval_steps_per_second": 2.823,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 23.56957087126138,
|
| 2317 |
+
"grad_norm": 2.4801974296569824,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.6311,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 23.56957087126138,
|
| 2324 |
+
"eval_accuracy": 0.8585083391102231,
|
| 2325 |
+
"eval_loss": 0.8391257524490356,
|
| 2326 |
+
"eval_runtime": 225.3955,
|
| 2327 |
+
"eval_samples_per_second": 90.658,
|
| 2328 |
+
"eval_steps_per_second": 2.835,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 23.73211963589077,
|
| 2333 |
+
"grad_norm": 2.4415576457977295,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.6333,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 23.73211963589077,
|
| 2340 |
+
"eval_accuracy": 0.8584229812472617,
|
| 2341 |
+
"eval_loss": 0.8474059104919434,
|
| 2342 |
+
"eval_runtime": 226.7984,
|
| 2343 |
+
"eval_samples_per_second": 90.098,
|
| 2344 |
+
"eval_steps_per_second": 2.817,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 23.894668400520157,
|
| 2349 |
+
"grad_norm": 2.5334153175354004,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.6291,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 23.894668400520157,
|
| 2356 |
+
"eval_accuracy": 0.8590801539747378,
|
| 2357 |
+
"eval_loss": 0.837401270866394,
|
| 2358 |
+
"eval_runtime": 225.1842,
|
| 2359 |
+
"eval_samples_per_second": 90.743,
|
| 2360 |
+
"eval_steps_per_second": 2.838,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 24.057217165149545,
|
| 2365 |
+
"grad_norm": 2.569241762161255,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.6255,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 24.057217165149545,
|
| 2372 |
+
"eval_accuracy": 0.8589290048653603,
|
| 2373 |
+
"eval_loss": 0.8332562446594238,
|
| 2374 |
+
"eval_runtime": 229.3684,
|
| 2375 |
+
"eval_samples_per_second": 89.088,
|
| 2376 |
+
"eval_steps_per_second": 2.786,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 24.219765929778934,
|
| 2381 |
+
"grad_norm": 2.526848316192627,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.6236,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 24.219765929778934,
|
| 2388 |
+
"eval_accuracy": 0.8597087528316705,
|
| 2389 |
+
"eval_loss": 0.848160445690155,
|
| 2390 |
+
"eval_runtime": 226.4635,
|
| 2391 |
+
"eval_samples_per_second": 90.231,
|
| 2392 |
+
"eval_steps_per_second": 2.822,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 24.382314694408322,
|
| 2397 |
+
"grad_norm": 2.655872106552124,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.624,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 24.382314694408322,
|
| 2404 |
+
"eval_accuracy": 0.8598462984123302,
|
| 2405 |
+
"eval_loss": 0.8461793065071106,
|
| 2406 |
+
"eval_runtime": 225.3124,
|
| 2407 |
+
"eval_samples_per_second": 90.692,
|
| 2408 |
+
"eval_steps_per_second": 2.836,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 24.54486345903771,
|
| 2413 |
+
"grad_norm": 2.109790325164795,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.6178,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 24.54486345903771,
|
| 2420 |
+
"eval_accuracy": 0.8595766101602678,
|
| 2421 |
+
"eval_loss": 0.8289109468460083,
|
| 2422 |
+
"eval_runtime": 226.5489,
|
| 2423 |
+
"eval_samples_per_second": 90.197,
|
| 2424 |
+
"eval_steps_per_second": 2.821,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 24.7074122236671,
|
| 2429 |
+
"grad_norm": 2.4060940742492676,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.6166,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 24.7074122236671,
|
| 2436 |
+
"eval_accuracy": 0.8606532868210031,
|
| 2437 |
+
"eval_loss": 0.8234532475471497,
|
| 2438 |
+
"eval_runtime": 225.3276,
|
| 2439 |
+
"eval_samples_per_second": 90.686,
|
| 2440 |
+
"eval_steps_per_second": 2.836,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 24.86996098829649,
|
| 2445 |
+
"grad_norm": 2.24013352394104,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.6247,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 24.86996098829649,
|
| 2452 |
+
"eval_accuracy": 0.8600803414966184,
|
| 2453 |
+
"eval_loss": 0.8327089548110962,
|
| 2454 |
+
"eval_runtime": 227.0917,
|
| 2455 |
+
"eval_samples_per_second": 89.981,
|
| 2456 |
+
"eval_steps_per_second": 2.814,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 25.03250975292588,
|
| 2461 |
+
"grad_norm": 2.370615005493164,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.624,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 25.03250975292588,
|
| 2468 |
+
"eval_accuracy": 0.8600759552685691,
|
| 2469 |
+
"eval_loss": 0.8379160761833191,
|
| 2470 |
+
"eval_runtime": 224.9721,
|
| 2471 |
+
"eval_samples_per_second": 90.829,
|
| 2472 |
+
"eval_steps_per_second": 2.84,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 25.195058517555267,
|
| 2477 |
+
"grad_norm": 2.192373514175415,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.6161,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 25.195058517555267,
|
| 2484 |
+
"eval_accuracy": 0.8597079760637308,
|
| 2485 |
+
"eval_loss": 0.8407602906227112,
|
| 2486 |
+
"eval_runtime": 225.6861,
|
| 2487 |
+
"eval_samples_per_second": 90.542,
|
| 2488 |
+
"eval_steps_per_second": 2.831,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 25.357607282184656,
|
| 2493 |
+
"grad_norm": 2.4751832485198975,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.6142,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 25.357607282184656,
|
| 2500 |
+
"eval_accuracy": 0.8608490074683309,
|
| 2501 |
+
"eval_loss": 0.8316988348960876,
|
| 2502 |
+
"eval_runtime": 225.4697,
|
| 2503 |
+
"eval_samples_per_second": 90.629,
|
| 2504 |
+
"eval_steps_per_second": 2.834,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 25.520156046814044,
|
| 2509 |
+
"grad_norm": 2.737602472305298,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.6193,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 25.520156046814044,
|
| 2516 |
+
"eval_accuracy": 0.860743624097801,
|
| 2517 |
+
"eval_loss": 0.8356801867485046,
|
| 2518 |
+
"eval_runtime": 226.5781,
|
| 2519 |
+
"eval_samples_per_second": 90.185,
|
| 2520 |
+
"eval_steps_per_second": 2.82,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 25.682704811443433,
|
| 2525 |
+
"grad_norm": 2.1464016437530518,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.6117,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 25.682704811443433,
|
| 2532 |
+
"eval_accuracy": 0.8607643217464414,
|
| 2533 |
+
"eval_loss": 0.8443505167961121,
|
| 2534 |
+
"eval_runtime": 225.3999,
|
| 2535 |
+
"eval_samples_per_second": 90.657,
|
| 2536 |
+
"eval_steps_per_second": 2.835,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 25.84525357607282,
|
| 2541 |
+
"grad_norm": 2.244690418243408,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.6155,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 25.84525357607282,
|
| 2548 |
+
"eval_accuracy": 0.8608062892739589,
|
| 2549 |
+
"eval_loss": 0.8309040665626526,
|
| 2550 |
+
"eval_runtime": 227.6535,
|
| 2551 |
+
"eval_samples_per_second": 89.759,
|
| 2552 |
+
"eval_steps_per_second": 2.807,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 26.00780234070221,
|
| 2557 |
+
"grad_norm": 2.2778093814849854,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.6159,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 26.00780234070221,
|
| 2564 |
+
"eval_accuracy": 0.860849372733563,
|
| 2565 |
+
"eval_loss": 0.8468282222747803,
|
| 2566 |
+
"eval_runtime": 226.4415,
|
| 2567 |
+
"eval_samples_per_second": 90.24,
|
| 2568 |
+
"eval_steps_per_second": 2.822,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 26.170351105331598,
|
| 2573 |
+
"grad_norm": 2.04994535446167,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.6095,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 26.170351105331598,
|
| 2580 |
+
"eval_accuracy": 0.8617138336966451,
|
| 2581 |
+
"eval_loss": 0.8285297155380249,
|
| 2582 |
+
"eval_runtime": 224.8524,
|
| 2583 |
+
"eval_samples_per_second": 90.877,
|
| 2584 |
+
"eval_steps_per_second": 2.842,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 26.33289986996099,
|
| 2589 |
+
"grad_norm": 2.437809944152832,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.6105,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 26.33289986996099,
|
| 2596 |
+
"eval_accuracy": 0.8610518577000663,
|
| 2597 |
+
"eval_loss": 0.8272110223770142,
|
| 2598 |
+
"eval_runtime": 225.6868,
|
| 2599 |
+
"eval_samples_per_second": 90.541,
|
| 2600 |
+
"eval_steps_per_second": 2.831,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 26.495448634590378,
|
| 2605 |
+
"grad_norm": 2.2250375747680664,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.6087,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 26.495448634590378,
|
| 2612 |
+
"eval_accuracy": 0.8626399458651031,
|
| 2613 |
+
"eval_loss": 0.812351644039154,
|
| 2614 |
+
"eval_runtime": 226.8612,
|
| 2615 |
+
"eval_samples_per_second": 90.073,
|
| 2616 |
+
"eval_steps_per_second": 2.817,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 26.657997399219767,
|
| 2621 |
+
"grad_norm": 2.060518980026245,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.6151,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 26.657997399219767,
|
| 2628 |
+
"eval_accuracy": 0.8616146515518304,
|
| 2629 |
+
"eval_loss": 0.8262282609939575,
|
| 2630 |
+
"eval_runtime": 226.3417,
|
| 2631 |
+
"eval_samples_per_second": 90.279,
|
| 2632 |
+
"eval_steps_per_second": 2.823,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 26.820546163849155,
|
| 2637 |
+
"grad_norm": 2.0575308799743652,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.6095,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 26.820546163849155,
|
| 2644 |
+
"eval_accuracy": 0.8622129077230878,
|
| 2645 |
+
"eval_loss": 0.8227624893188477,
|
| 2646 |
+
"eval_runtime": 228.246,
|
| 2647 |
+
"eval_samples_per_second": 89.526,
|
| 2648 |
+
"eval_steps_per_second": 2.8,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 26.983094928478543,
|
| 2653 |
+
"grad_norm": 2.253310203552246,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.6,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 26.983094928478543,
|
| 2660 |
+
"eval_accuracy": 0.8627496487013281,
|
| 2661 |
+
"eval_loss": 0.8195393085479736,
|
| 2662 |
+
"eval_runtime": 226.7315,
|
| 2663 |
+
"eval_samples_per_second": 90.124,
|
| 2664 |
+
"eval_steps_per_second": 2.818,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 27.14564369310793,
|
| 2669 |
+
"grad_norm": 2.459129810333252,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.6013,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 27.14564369310793,
|
| 2676 |
+
"eval_accuracy": 0.8624795555933904,
|
| 2677 |
+
"eval_loss": 0.8182792663574219,
|
| 2678 |
+
"eval_runtime": 225.0241,
|
| 2679 |
+
"eval_samples_per_second": 90.808,
|
| 2680 |
+
"eval_steps_per_second": 2.84,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 27.30819245773732,
|
| 2685 |
+
"grad_norm": 2.1593105792999268,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.6001,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 27.30819245773732,
|
| 2692 |
+
"eval_accuracy": 0.8627880443457693,
|
| 2693 |
+
"eval_loss": 0.8216105103492737,
|
| 2694 |
+
"eval_runtime": 224.8609,
|
| 2695 |
+
"eval_samples_per_second": 90.874,
|
| 2696 |
+
"eval_steps_per_second": 2.842,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 27.47074122236671,
|
| 2701 |
+
"grad_norm": 2.086055040359497,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.6013,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 27.47074122236671,
|
| 2708 |
+
"eval_accuracy": 0.8621898394825607,
|
| 2709 |
+
"eval_loss": 0.8210575580596924,
|
| 2710 |
+
"eval_runtime": 226.231,
|
| 2711 |
+
"eval_samples_per_second": 90.324,
|
| 2712 |
+
"eval_steps_per_second": 2.825,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 27.6332899869961,
|
| 2717 |
+
"grad_norm": 2.5327186584472656,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.6058,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 27.6332899869961,
|
| 2724 |
+
"eval_accuracy": 0.8627229258499225,
|
| 2725 |
+
"eval_loss": 0.8185027241706848,
|
| 2726 |
+
"eval_runtime": 226.1064,
|
| 2727 |
+
"eval_samples_per_second": 90.373,
|
| 2728 |
+
"eval_steps_per_second": 2.826,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 27.79583875162549,
|
| 2733 |
+
"grad_norm": 2.3839502334594727,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.6042,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 27.79583875162549,
|
| 2740 |
+
"eval_accuracy": 0.8627582312704439,
|
| 2741 |
+
"eval_loss": 0.8231362104415894,
|
| 2742 |
+
"eval_runtime": 228.8409,
|
| 2743 |
+
"eval_samples_per_second": 89.293,
|
| 2744 |
+
"eval_steps_per_second": 2.792,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 27.958387516254877,
|
| 2749 |
+
"grad_norm": 2.195699691772461,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.5997,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 27.958387516254877,
|
| 2756 |
+
"eval_accuracy": 0.8631414086547167,
|
| 2757 |
+
"eval_loss": 0.8138222694396973,
|
| 2758 |
+
"eval_runtime": 225.1659,
|
| 2759 |
+
"eval_samples_per_second": 90.751,
|
| 2760 |
+
"eval_steps_per_second": 2.838,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 28.120936280884266,
|
| 2765 |
+
"grad_norm": 2.942133665084839,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.5976,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 28.120936280884266,
|
| 2772 |
+
"eval_accuracy": 0.8629241152120203,
|
| 2773 |
+
"eval_loss": 0.8277115225791931,
|
| 2774 |
+
"eval_runtime": 226.384,
|
| 2775 |
+
"eval_samples_per_second": 90.263,
|
| 2776 |
+
"eval_steps_per_second": 2.823,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 28.283485045513654,
|
| 2781 |
+
"grad_norm": 2.2104339599609375,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.6005,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 28.283485045513654,
|
| 2788 |
+
"eval_accuracy": 0.8635328035684335,
|
| 2789 |
+
"eval_loss": 0.8249954581260681,
|
| 2790 |
+
"eval_runtime": 225.8607,
|
| 2791 |
+
"eval_samples_per_second": 90.472,
|
| 2792 |
+
"eval_steps_per_second": 2.829,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 28.446033810143042,
|
| 2797 |
+
"grad_norm": 2.4792349338531494,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.5964,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 28.446033810143042,
|
| 2804 |
+
"eval_accuracy": 0.8631331114250733,
|
| 2805 |
+
"eval_loss": 0.8169623613357544,
|
| 2806 |
+
"eval_runtime": 224.4723,
|
| 2807 |
+
"eval_samples_per_second": 91.031,
|
| 2808 |
+
"eval_steps_per_second": 2.847,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 28.60858257477243,
|
| 2813 |
+
"grad_norm": 2.2584621906280518,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.5978,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 28.60858257477243,
|
| 2820 |
+
"eval_accuracy": 0.8637502162788443,
|
| 2821 |
+
"eval_loss": 0.8240325450897217,
|
| 2822 |
+
"eval_runtime": 226.0844,
|
| 2823 |
+
"eval_samples_per_second": 90.382,
|
| 2824 |
+
"eval_steps_per_second": 2.826,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 28.77113133940182,
|
| 2829 |
+
"grad_norm": 2.4756360054016113,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.5933,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 28.77113133940182,
|
| 2836 |
+
"eval_accuracy": 0.863286343484686,
|
| 2837 |
+
"eval_loss": 0.821010947227478,
|
| 2838 |
+
"eval_runtime": 233.6708,
|
| 2839 |
+
"eval_samples_per_second": 87.448,
|
| 2840 |
+
"eval_steps_per_second": 2.735,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 28.93368010403121,
|
| 2845 |
+
"grad_norm": 2.4466359615325928,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.595,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 28.93368010403121,
|
| 2852 |
+
"eval_accuracy": 0.8634251119288006,
|
| 2853 |
+
"eval_loss": 0.821042001247406,
|
| 2854 |
+
"eval_runtime": 225.7222,
|
| 2855 |
+
"eval_samples_per_second": 90.527,
|
| 2856 |
+
"eval_steps_per_second": 2.831,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 29.0962288686606,
|
| 2861 |
+
"grad_norm": 2.5361807346343994,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.5941,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 29.0962288686606,
|
| 2868 |
+
"eval_accuracy": 0.8634033694143568,
|
| 2869 |
+
"eval_loss": 0.8251069188117981,
|
| 2870 |
+
"eval_runtime": 226.448,
|
| 2871 |
+
"eval_samples_per_second": 90.237,
|
| 2872 |
+
"eval_steps_per_second": 2.822,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 29.258777633289988,
|
| 2877 |
+
"grad_norm": 2.5238239765167236,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.5988,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 29.258777633289988,
|
| 2884 |
+
"eval_accuracy": 0.8638910896653752,
|
| 2885 |
+
"eval_loss": 0.8146407604217529,
|
| 2886 |
+
"eval_runtime": 222.0422,
|
| 2887 |
+
"eval_samples_per_second": 92.028,
|
| 2888 |
+
"eval_steps_per_second": 2.878,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 29.421326397919376,
|
| 2893 |
+
"grad_norm": 2.291041851043701,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.5923,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 29.421326397919376,
|
| 2900 |
+
"eval_accuracy": 0.8637874627536761,
|
| 2901 |
+
"eval_loss": 0.8178155422210693,
|
| 2902 |
+
"eval_runtime": 221.0933,
|
| 2903 |
+
"eval_samples_per_second": 92.423,
|
| 2904 |
+
"eval_steps_per_second": 2.89,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 29.583875162548765,
|
| 2909 |
+
"grad_norm": 2.601836681365967,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.5887,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 29.583875162548765,
|
| 2916 |
+
"eval_accuracy": 0.864438274609619,
|
| 2917 |
+
"eval_loss": 0.8162264823913574,
|
| 2918 |
+
"eval_runtime": 228.6953,
|
| 2919 |
+
"eval_samples_per_second": 89.35,
|
| 2920 |
+
"eval_steps_per_second": 2.794,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 29.746423927178153,
|
| 2925 |
+
"grad_norm": 2.56785249710083,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.5833,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 29.746423927178153,
|
| 2932 |
+
"eval_accuracy": 0.8640183304537222,
|
| 2933 |
+
"eval_loss": 0.8157439827919006,
|
| 2934 |
+
"eval_runtime": 224.1585,
|
| 2935 |
+
"eval_samples_per_second": 91.159,
|
| 2936 |
+
"eval_steps_per_second": 2.851,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 29.90897269180754,
|
| 2941 |
+
"grad_norm": 2.409670114517212,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.5951,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 29.90897269180754,
|
| 2948 |
+
"eval_accuracy": 0.8643441985332161,
|
| 2949 |
+
"eval_loss": 0.812107264995575,
|
| 2950 |
+
"eval_runtime": 221.533,
|
| 2951 |
+
"eval_samples_per_second": 92.239,
|
| 2952 |
+
"eval_steps_per_second": 2.884,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 30.07152145643693,
|
| 2957 |
+
"grad_norm": 2.19612717628479,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.5928,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 30.07152145643693,
|
| 2964 |
+
"eval_accuracy": 0.8640605097561019,
|
| 2965 |
+
"eval_loss": 0.8203216791152954,
|
| 2966 |
+
"eval_runtime": 221.1524,
|
| 2967 |
+
"eval_samples_per_second": 92.398,
|
| 2968 |
+
"eval_steps_per_second": 2.889,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 30.234070221066318,
|
| 2973 |
+
"grad_norm": 2.249537944793701,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.5878,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 30.234070221066318,
|
| 2980 |
+
"eval_accuracy": 0.8644494113060075,
|
| 2981 |
+
"eval_loss": 0.8103818893432617,
|
| 2982 |
+
"eval_runtime": 221.6978,
|
| 2983 |
+
"eval_samples_per_second": 92.17,
|
| 2984 |
+
"eval_steps_per_second": 2.882,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 30.39661898569571,
|
| 2989 |
+
"grad_norm": 2.63948917388916,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.5831,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 30.39661898569571,
|
| 2996 |
+
"eval_accuracy": 0.8646731645828304,
|
| 2997 |
+
"eval_loss": 0.8149096369743347,
|
| 2998 |
+
"eval_runtime": 231.6244,
|
| 2999 |
+
"eval_samples_per_second": 88.22,
|
| 3000 |
+
"eval_steps_per_second": 2.759,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 30.5591677503251,
|
| 3005 |
+
"grad_norm": 2.257843494415283,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.5922,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 30.5591677503251,
|
| 3012 |
+
"eval_accuracy": 0.8642099290677292,
|
| 3013 |
+
"eval_loss": 0.8136395215988159,
|
| 3014 |
+
"eval_runtime": 227.8537,
|
| 3015 |
+
"eval_samples_per_second": 89.68,
|
| 3016 |
+
"eval_steps_per_second": 2.804,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 30.721716514954487,
|
| 3021 |
+
"grad_norm": 2.54514479637146,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.5834,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 30.721716514954487,
|
| 3028 |
+
"eval_accuracy": 0.8647608401845014,
|
| 3029 |
+
"eval_loss": 0.8055408596992493,
|
| 3030 |
+
"eval_runtime": 222.2677,
|
| 3031 |
+
"eval_samples_per_second": 91.934,
|
| 3032 |
+
"eval_steps_per_second": 2.875,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 30.884265279583875,
|
| 3037 |
+
"grad_norm": 2.5504415035247803,
|
| 3038 |
+
"learning_rate": 5e-06,
|
| 3039 |
+
"loss": 0.5851,
|
| 3040 |
+
"step": 95000
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 30.884265279583875,
|
| 3044 |
+
"eval_accuracy": 0.8655900321339209,
|
| 3045 |
+
"eval_loss": 0.8102414011955261,
|
| 3046 |
+
"eval_runtime": 222.3833,
|
| 3047 |
+
"eval_samples_per_second": 91.886,
|
| 3048 |
+
"eval_steps_per_second": 2.873,
|
| 3049 |
+
"step": 95000
|
| 3050 |
+
},
|
| 3051 |
+
{
|
| 3052 |
+
"epoch": 31.046814044213264,
|
| 3053 |
+
"grad_norm": 2.5223162174224854,
|
| 3054 |
+
"learning_rate": 4.5e-06,
|
| 3055 |
+
"loss": 0.5858,
|
| 3056 |
+
"step": 95500
|
| 3057 |
+
},
|
| 3058 |
+
{
|
| 3059 |
+
"epoch": 31.046814044213264,
|
| 3060 |
+
"eval_accuracy": 0.8648693759071118,
|
| 3061 |
+
"eval_loss": 0.8096651434898376,
|
| 3062 |
+
"eval_runtime": 222.4149,
|
| 3063 |
+
"eval_samples_per_second": 91.873,
|
| 3064 |
+
"eval_steps_per_second": 2.873,
|
| 3065 |
+
"step": 95500
|
| 3066 |
+
},
|
| 3067 |
+
{
|
| 3068 |
+
"epoch": 31.209362808842652,
|
| 3069 |
+
"grad_norm": 2.704538583755493,
|
| 3070 |
+
"learning_rate": 4.000000000000001e-06,
|
| 3071 |
+
"loss": 0.5854,
|
| 3072 |
+
"step": 96000
|
| 3073 |
+
},
|
| 3074 |
+
{
|
| 3075 |
+
"epoch": 31.209362808842652,
|
| 3076 |
+
"eval_accuracy": 0.8643289659912363,
|
| 3077 |
+
"eval_loss": 0.8139033317565918,
|
| 3078 |
+
"eval_runtime": 222.3309,
|
| 3079 |
+
"eval_samples_per_second": 91.908,
|
| 3080 |
+
"eval_steps_per_second": 2.874,
|
| 3081 |
+
"step": 96000
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 31.37191157347204,
|
| 3085 |
+
"grad_norm": 2.3286654949188232,
|
| 3086 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 3087 |
+
"loss": 0.5809,
|
| 3088 |
+
"step": 96500
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 31.37191157347204,
|
| 3092 |
+
"eval_accuracy": 0.8655915682628923,
|
| 3093 |
+
"eval_loss": 0.8057557940483093,
|
| 3094 |
+
"eval_runtime": 222.7618,
|
| 3095 |
+
"eval_samples_per_second": 91.73,
|
| 3096 |
+
"eval_steps_per_second": 2.869,
|
| 3097 |
+
"step": 96500
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 31.53446033810143,
|
| 3101 |
+
"grad_norm": 2.572178363800049,
|
| 3102 |
+
"learning_rate": 3e-06,
|
| 3103 |
+
"loss": 0.5846,
|
| 3104 |
+
"step": 97000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 31.53446033810143,
|
| 3108 |
+
"eval_accuracy": 0.8653023190042686,
|
| 3109 |
+
"eval_loss": 0.8064507842063904,
|
| 3110 |
+
"eval_runtime": 223.3613,
|
| 3111 |
+
"eval_samples_per_second": 91.484,
|
| 3112 |
+
"eval_steps_per_second": 2.861,
|
| 3113 |
+
"step": 97000
|
| 3114 |
+
},
|
| 3115 |
+
{
|
| 3116 |
+
"epoch": 31.69700910273082,
|
| 3117 |
+
"grad_norm": 2.4184162616729736,
|
| 3118 |
+
"learning_rate": 2.5e-06,
|
| 3119 |
+
"loss": 0.585,
|
| 3120 |
+
"step": 97500
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"epoch": 31.69700910273082,
|
| 3124 |
+
"eval_accuracy": 0.8657269498564959,
|
| 3125 |
+
"eval_loss": 0.7961297631263733,
|
| 3126 |
+
"eval_runtime": 225.1117,
|
| 3127 |
+
"eval_samples_per_second": 90.773,
|
| 3128 |
+
"eval_steps_per_second": 2.839,
|
| 3129 |
+
"step": 97500
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 31.85955786736021,
|
| 3133 |
+
"grad_norm": 2.121875047683716,
|
| 3134 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 3135 |
+
"loss": 0.588,
|
| 3136 |
+
"step": 98000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 31.85955786736021,
|
| 3140 |
+
"eval_accuracy": 0.8654759221390312,
|
| 3141 |
+
"eval_loss": 0.7994445562362671,
|
| 3142 |
+
"eval_runtime": 222.5184,
|
| 3143 |
+
"eval_samples_per_second": 91.831,
|
| 3144 |
+
"eval_steps_per_second": 2.872,
|
| 3145 |
+
"step": 98000
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 32.022106631989594,
|
| 3149 |
+
"grad_norm": 2.2691831588745117,
|
| 3150 |
+
"learning_rate": 1.5e-06,
|
| 3151 |
+
"loss": 0.58,
|
| 3152 |
+
"step": 98500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 32.022106631989594,
|
| 3156 |
+
"eval_accuracy": 0.8651082523885062,
|
| 3157 |
+
"eval_loss": 0.8115944266319275,
|
| 3158 |
+
"eval_runtime": 222.6443,
|
| 3159 |
+
"eval_samples_per_second": 91.779,
|
| 3160 |
+
"eval_steps_per_second": 2.87,
|
| 3161 |
+
"step": 98500
|
| 3162 |
+
},
|
| 3163 |
+
{
|
| 3164 |
+
"epoch": 32.184655396618986,
|
| 3165 |
+
"grad_norm": 2.047391891479492,
|
| 3166 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 3167 |
+
"loss": 0.5776,
|
| 3168 |
+
"step": 99000
|
| 3169 |
+
},
|
| 3170 |
+
{
|
| 3171 |
+
"epoch": 32.184655396618986,
|
| 3172 |
+
"eval_accuracy": 0.8659600453633429,
|
| 3173 |
+
"eval_loss": 0.8002920150756836,
|
| 3174 |
+
"eval_runtime": 223.1406,
|
| 3175 |
+
"eval_samples_per_second": 91.575,
|
| 3176 |
+
"eval_steps_per_second": 2.864,
|
| 3177 |
+
"step": 99000
|
| 3178 |
+
},
|
| 3179 |
+
{
|
| 3180 |
+
"epoch": 32.34720416124838,
|
| 3181 |
+
"grad_norm": 2.4214348793029785,
|
| 3182 |
+
"learning_rate": 5.000000000000001e-07,
|
| 3183 |
+
"loss": 0.581,
|
| 3184 |
+
"step": 99500
|
| 3185 |
+
},
|
| 3186 |
+
{
|
| 3187 |
+
"epoch": 32.34720416124838,
|
| 3188 |
+
"eval_accuracy": 0.8655375271936185,
|
| 3189 |
+
"eval_loss": 0.7950243949890137,
|
| 3190 |
+
"eval_runtime": 240.505,
|
| 3191 |
+
"eval_samples_per_second": 84.963,
|
| 3192 |
+
"eval_steps_per_second": 2.657,
|
| 3193 |
+
"step": 99500
|
| 3194 |
+
}
|
| 3195 |
+
],
|
| 3196 |
+
"logging_steps": 500,
|
| 3197 |
+
"max_steps": 100000,
|
| 3198 |
+
"num_input_tokens_seen": 0,
|
| 3199 |
+
"num_train_epochs": 33,
|
| 3200 |
+
"save_steps": 500,
|
| 3201 |
+
"stateful_callbacks": {
|
| 3202 |
+
"TrainerControl": {
|
| 3203 |
+
"args": {
|
| 3204 |
+
"should_epoch_stop": false,
|
| 3205 |
+
"should_evaluate": false,
|
| 3206 |
+
"should_log": false,
|
| 3207 |
+
"should_save": true,
|
| 3208 |
+
"should_training_stop": false
|
| 3209 |
+
},
|
| 3210 |
+
"attributes": {}
|
| 3211 |
+
}
|
| 3212 |
+
},
|
| 3213 |
+
"total_flos": 8.401908931210772e+17,
|
| 3214 |
+
"train_batch_size": 16,
|
| 3215 |
+
"trial_name": null,
|
| 3216 |
+
"trial_params": null
|
| 3217 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87c347a80e831249008f8fb43269447bcd3741f34dd1b039d8871191a9960cf7
|
| 3 |
+
size 5176
|