Uploading checkpoint-94500 for xlm-r - amh-ethi
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 +3057 -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: amh-Ethi
|
| 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 |
+
# amh-Ethi
|
| 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: 0.8454
|
| 22 |
+
- Accuracy: 0.8217
|
| 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:c98e7b3fc67a69f569db27136eb24024a9a98c81ed11bb39498884e9daeae135
|
| 3 |
+
size 1113205088
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8fa0035f2ebc4bfa204c3d30464ac50943a99ff1fa45dc9b895de4c7f736023
|
| 3 |
+
size 2226530490
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b96dee0847b49d2bd5a823ac3e9830460010eac9e2549102ff9ed51de023110c
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8651376b9472f3341e0744a1514bf187c0037eee4ec72a792fb496a7a83efa5a
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25106ae1ffc1c5bc8bdb8df44004f07b4bcb5172b622a1efa6ecdd9bb9f1257e
|
| 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 +1,3057 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8419870138168335,
|
| 3 |
+
"best_model_checkpoint": "./model_fine-tune/glot/xlm-r/amh-Ethi/checkpoint-94500",
|
| 4 |
+
"epoch": 121.46529562982005,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 94500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.6426735218508998,
|
| 13 |
+
"grad_norm": 4.40508508682251,
|
| 14 |
+
"learning_rate": 9.95e-05,
|
| 15 |
+
"loss": 1.6805,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.6426735218508998,
|
| 20 |
+
"eval_accuracy": 0.6904924931069853,
|
| 21 |
+
"eval_loss": 1.4787547588348389,
|
| 22 |
+
"eval_runtime": 34.6674,
|
| 23 |
+
"eval_samples_per_second": 179.506,
|
| 24 |
+
"eval_steps_per_second": 5.625,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.2853470437017995,
|
| 29 |
+
"grad_norm": 3.9189295768737793,
|
| 30 |
+
"learning_rate": 9.900000000000001e-05,
|
| 31 |
+
"loss": 1.5356,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.2853470437017995,
|
| 36 |
+
"eval_accuracy": 0.7092839911070465,
|
| 37 |
+
"eval_loss": 1.3684136867523193,
|
| 38 |
+
"eval_runtime": 35.1037,
|
| 39 |
+
"eval_samples_per_second": 177.275,
|
| 40 |
+
"eval_steps_per_second": 5.555,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.9280205655526992,
|
| 45 |
+
"grad_norm": 4.308630466461182,
|
| 46 |
+
"learning_rate": 9.850000000000001e-05,
|
| 47 |
+
"loss": 1.4583,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.9280205655526992,
|
| 52 |
+
"eval_accuracy": 0.7174822425497339,
|
| 53 |
+
"eval_loss": 1.3316432237625122,
|
| 54 |
+
"eval_runtime": 32.7687,
|
| 55 |
+
"eval_samples_per_second": 189.907,
|
| 56 |
+
"eval_steps_per_second": 5.951,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.570694087403599,
|
| 61 |
+
"grad_norm": 3.4791815280914307,
|
| 62 |
+
"learning_rate": 9.8e-05,
|
| 63 |
+
"loss": 1.4029,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.570694087403599,
|
| 68 |
+
"eval_accuracy": 0.7272827658735376,
|
| 69 |
+
"eval_loss": 1.280930519104004,
|
| 70 |
+
"eval_runtime": 35.7018,
|
| 71 |
+
"eval_samples_per_second": 174.305,
|
| 72 |
+
"eval_steps_per_second": 5.462,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 3.2133676092544987,
|
| 77 |
+
"grad_norm": 3.705988645553589,
|
| 78 |
+
"learning_rate": 9.75e-05,
|
| 79 |
+
"loss": 1.3485,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.2133676092544987,
|
| 84 |
+
"eval_accuracy": 0.7312191932269085,
|
| 85 |
+
"eval_loss": 1.2635687589645386,
|
| 86 |
+
"eval_runtime": 34.9687,
|
| 87 |
+
"eval_samples_per_second": 177.959,
|
| 88 |
+
"eval_steps_per_second": 5.576,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.8560411311053984,
|
| 93 |
+
"grad_norm": 3.63777232170105,
|
| 94 |
+
"learning_rate": 9.7e-05,
|
| 95 |
+
"loss": 1.318,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.8560411311053984,
|
| 100 |
+
"eval_accuracy": 0.7356960949650779,
|
| 101 |
+
"eval_loss": 1.2287126779556274,
|
| 102 |
+
"eval_runtime": 35.1013,
|
| 103 |
+
"eval_samples_per_second": 177.287,
|
| 104 |
+
"eval_steps_per_second": 5.555,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 4.4987146529562985,
|
| 109 |
+
"grad_norm": 3.065805673599243,
|
| 110 |
+
"learning_rate": 9.65e-05,
|
| 111 |
+
"loss": 1.2853,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.4987146529562985,
|
| 116 |
+
"eval_accuracy": 0.7411218708879311,
|
| 117 |
+
"eval_loss": 1.2096679210662842,
|
| 118 |
+
"eval_runtime": 35.3152,
|
| 119 |
+
"eval_samples_per_second": 176.213,
|
| 120 |
+
"eval_steps_per_second": 5.522,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.141388174807198,
|
| 125 |
+
"grad_norm": 3.083160161972046,
|
| 126 |
+
"learning_rate": 9.6e-05,
|
| 127 |
+
"loss": 1.2579,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.141388174807198,
|
| 132 |
+
"eval_accuracy": 0.743514192750392,
|
| 133 |
+
"eval_loss": 1.1946351528167725,
|
| 134 |
+
"eval_runtime": 35.3519,
|
| 135 |
+
"eval_samples_per_second": 176.03,
|
| 136 |
+
"eval_steps_per_second": 5.516,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.784061696658098,
|
| 141 |
+
"grad_norm": 3.1521048545837402,
|
| 142 |
+
"learning_rate": 9.55e-05,
|
| 143 |
+
"loss": 1.225,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 5.784061696658098,
|
| 148 |
+
"eval_accuracy": 0.7478054754473986,
|
| 149 |
+
"eval_loss": 1.1656205654144287,
|
| 150 |
+
"eval_runtime": 35.2967,
|
| 151 |
+
"eval_samples_per_second": 176.305,
|
| 152 |
+
"eval_steps_per_second": 5.525,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 6.426735218508997,
|
| 157 |
+
"grad_norm": 2.947038412094116,
|
| 158 |
+
"learning_rate": 9.5e-05,
|
| 159 |
+
"loss": 1.2156,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 6.426735218508997,
|
| 164 |
+
"eval_accuracy": 0.7494688799077729,
|
| 165 |
+
"eval_loss": 1.1547260284423828,
|
| 166 |
+
"eval_runtime": 35.244,
|
| 167 |
+
"eval_samples_per_second": 176.569,
|
| 168 |
+
"eval_steps_per_second": 5.533,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 7.069408740359897,
|
| 173 |
+
"grad_norm": 3.245880603790283,
|
| 174 |
+
"learning_rate": 9.449999999999999e-05,
|
| 175 |
+
"loss": 1.1878,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 7.069408740359897,
|
| 180 |
+
"eval_accuracy": 0.7533389855345517,
|
| 181 |
+
"eval_loss": 1.1394718885421753,
|
| 182 |
+
"eval_runtime": 34.1955,
|
| 183 |
+
"eval_samples_per_second": 181.983,
|
| 184 |
+
"eval_steps_per_second": 5.703,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.712082262210797,
|
| 189 |
+
"grad_norm": 3.02457857131958,
|
| 190 |
+
"learning_rate": 9.4e-05,
|
| 191 |
+
"loss": 1.1713,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 7.712082262210797,
|
| 196 |
+
"eval_accuracy": 0.7548033682748899,
|
| 197 |
+
"eval_loss": 1.136972188949585,
|
| 198 |
+
"eval_runtime": 33.9827,
|
| 199 |
+
"eval_samples_per_second": 183.123,
|
| 200 |
+
"eval_steps_per_second": 5.738,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 8.354755784061696,
|
| 205 |
+
"grad_norm": 2.8089816570281982,
|
| 206 |
+
"learning_rate": 9.350000000000001e-05,
|
| 207 |
+
"loss": 1.1587,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 8.354755784061696,
|
| 212 |
+
"eval_accuracy": 0.7558409318256807,
|
| 213 |
+
"eval_loss": 1.1281076669692993,
|
| 214 |
+
"eval_runtime": 34.9074,
|
| 215 |
+
"eval_samples_per_second": 178.272,
|
| 216 |
+
"eval_steps_per_second": 5.586,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.997429305912597,
|
| 221 |
+
"grad_norm": 2.808964729309082,
|
| 222 |
+
"learning_rate": 9.300000000000001e-05,
|
| 223 |
+
"loss": 1.1416,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.997429305912597,
|
| 228 |
+
"eval_accuracy": 0.7600179626094105,
|
| 229 |
+
"eval_loss": 1.107399821281433,
|
| 230 |
+
"eval_runtime": 34.0824,
|
| 231 |
+
"eval_samples_per_second": 182.587,
|
| 232 |
+
"eval_steps_per_second": 5.721,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 9.640102827763496,
|
| 237 |
+
"grad_norm": 2.8558380603790283,
|
| 238 |
+
"learning_rate": 9.250000000000001e-05,
|
| 239 |
+
"loss": 1.128,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 9.640102827763496,
|
| 244 |
+
"eval_accuracy": 0.7628805020537848,
|
| 245 |
+
"eval_loss": 1.0930116176605225,
|
| 246 |
+
"eval_runtime": 33.4368,
|
| 247 |
+
"eval_samples_per_second": 186.112,
|
| 248 |
+
"eval_steps_per_second": 5.832,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 10.282776349614396,
|
| 253 |
+
"grad_norm": 2.9800262451171875,
|
| 254 |
+
"learning_rate": 9.200000000000001e-05,
|
| 255 |
+
"loss": 1.1213,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 10.282776349614396,
|
| 260 |
+
"eval_accuracy": 0.7627036601926251,
|
| 261 |
+
"eval_loss": 1.0944606065750122,
|
| 262 |
+
"eval_runtime": 34.4461,
|
| 263 |
+
"eval_samples_per_second": 180.659,
|
| 264 |
+
"eval_steps_per_second": 5.661,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 10.925449871465295,
|
| 269 |
+
"grad_norm": 3.040515422821045,
|
| 270 |
+
"learning_rate": 9.15e-05,
|
| 271 |
+
"loss": 1.1075,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 10.925449871465295,
|
| 276 |
+
"eval_accuracy": 0.764068771058977,
|
| 277 |
+
"eval_loss": 1.0873337984085083,
|
| 278 |
+
"eval_runtime": 32.7071,
|
| 279 |
+
"eval_samples_per_second": 190.265,
|
| 280 |
+
"eval_steps_per_second": 5.962,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 11.568123393316196,
|
| 285 |
+
"grad_norm": 2.973078966140747,
|
| 286 |
+
"learning_rate": 9.1e-05,
|
| 287 |
+
"loss": 1.0912,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 11.568123393316196,
|
| 292 |
+
"eval_accuracy": 0.7658058887449325,
|
| 293 |
+
"eval_loss": 1.0768815279006958,
|
| 294 |
+
"eval_runtime": 32.6454,
|
| 295 |
+
"eval_samples_per_second": 190.624,
|
| 296 |
+
"eval_steps_per_second": 5.973,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 12.210796915167094,
|
| 301 |
+
"grad_norm": 3.5373220443725586,
|
| 302 |
+
"learning_rate": 9.05e-05,
|
| 303 |
+
"loss": 1.085,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 12.210796915167094,
|
| 308 |
+
"eval_accuracy": 0.7658247272593731,
|
| 309 |
+
"eval_loss": 1.086098074913025,
|
| 310 |
+
"eval_runtime": 32.662,
|
| 311 |
+
"eval_samples_per_second": 190.527,
|
| 312 |
+
"eval_steps_per_second": 5.97,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 12.853470437017995,
|
| 317 |
+
"grad_norm": 5.5106964111328125,
|
| 318 |
+
"learning_rate": 9e-05,
|
| 319 |
+
"loss": 1.0746,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 12.853470437017995,
|
| 324 |
+
"eval_accuracy": 0.7682209938962093,
|
| 325 |
+
"eval_loss": 1.0635614395141602,
|
| 326 |
+
"eval_runtime": 32.6593,
|
| 327 |
+
"eval_samples_per_second": 190.543,
|
| 328 |
+
"eval_steps_per_second": 5.971,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.496143958868895,
|
| 333 |
+
"grad_norm": 3.238605499267578,
|
| 334 |
+
"learning_rate": 8.950000000000001e-05,
|
| 335 |
+
"loss": 1.0531,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 13.496143958868895,
|
| 340 |
+
"eval_accuracy": 0.7686876543500863,
|
| 341 |
+
"eval_loss": 1.0645390748977661,
|
| 342 |
+
"eval_runtime": 32.6258,
|
| 343 |
+
"eval_samples_per_second": 190.739,
|
| 344 |
+
"eval_steps_per_second": 5.977,
|
| 345 |
+
"step": 10500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 14.138817480719794,
|
| 349 |
+
"grad_norm": 2.7707929611206055,
|
| 350 |
+
"learning_rate": 8.900000000000001e-05,
|
| 351 |
+
"loss": 1.0545,
|
| 352 |
+
"step": 11000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 14.138817480719794,
|
| 356 |
+
"eval_accuracy": 0.7713913141190644,
|
| 357 |
+
"eval_loss": 1.0523449182510376,
|
| 358 |
+
"eval_runtime": 33.0121,
|
| 359 |
+
"eval_samples_per_second": 188.507,
|
| 360 |
+
"eval_steps_per_second": 5.907,
|
| 361 |
+
"step": 11000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 14.781491002570695,
|
| 365 |
+
"grad_norm": 3.1346287727355957,
|
| 366 |
+
"learning_rate": 8.850000000000001e-05,
|
| 367 |
+
"loss": 1.042,
|
| 368 |
+
"step": 11500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 14.781491002570695,
|
| 372 |
+
"eval_accuracy": 0.7711833853943258,
|
| 373 |
+
"eval_loss": 1.0591264963150024,
|
| 374 |
+
"eval_runtime": 32.5788,
|
| 375 |
+
"eval_samples_per_second": 191.014,
|
| 376 |
+
"eval_steps_per_second": 5.985,
|
| 377 |
+
"step": 11500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 15.424164524421593,
|
| 381 |
+
"grad_norm": 2.98738169670105,
|
| 382 |
+
"learning_rate": 8.800000000000001e-05,
|
| 383 |
+
"loss": 1.0327,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 15.424164524421593,
|
| 388 |
+
"eval_accuracy": 0.7727139790786457,
|
| 389 |
+
"eval_loss": 1.0433602333068848,
|
| 390 |
+
"eval_runtime": 32.6912,
|
| 391 |
+
"eval_samples_per_second": 190.357,
|
| 392 |
+
"eval_steps_per_second": 5.965,
|
| 393 |
+
"step": 12000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 16.066838046272494,
|
| 397 |
+
"grad_norm": 2.7987964153289795,
|
| 398 |
+
"learning_rate": 8.75e-05,
|
| 399 |
+
"loss": 1.0226,
|
| 400 |
+
"step": 12500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 16.066838046272494,
|
| 404 |
+
"eval_accuracy": 0.7744916530689092,
|
| 405 |
+
"eval_loss": 1.0427190065383911,
|
| 406 |
+
"eval_runtime": 32.7105,
|
| 407 |
+
"eval_samples_per_second": 190.245,
|
| 408 |
+
"eval_steps_per_second": 5.961,
|
| 409 |
+
"step": 12500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 16.709511568123393,
|
| 413 |
+
"grad_norm": 2.927727699279785,
|
| 414 |
+
"learning_rate": 8.7e-05,
|
| 415 |
+
"loss": 1.0164,
|
| 416 |
+
"step": 13000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 16.709511568123393,
|
| 420 |
+
"eval_accuracy": 0.7755782307589745,
|
| 421 |
+
"eval_loss": 1.033629059791565,
|
| 422 |
+
"eval_runtime": 32.6997,
|
| 423 |
+
"eval_samples_per_second": 190.307,
|
| 424 |
+
"eval_steps_per_second": 5.963,
|
| 425 |
+
"step": 13000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 17.35218508997429,
|
| 429 |
+
"grad_norm": 2.902130603790283,
|
| 430 |
+
"learning_rate": 8.65e-05,
|
| 431 |
+
"loss": 1.012,
|
| 432 |
+
"step": 13500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 17.35218508997429,
|
| 436 |
+
"eval_accuracy": 0.7768737795368466,
|
| 437 |
+
"eval_loss": 1.024613380432129,
|
| 438 |
+
"eval_runtime": 32.6543,
|
| 439 |
+
"eval_samples_per_second": 190.572,
|
| 440 |
+
"eval_steps_per_second": 5.972,
|
| 441 |
+
"step": 13500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 17.994858611825194,
|
| 445 |
+
"grad_norm": 2.6647133827209473,
|
| 446 |
+
"learning_rate": 8.6e-05,
|
| 447 |
+
"loss": 0.9992,
|
| 448 |
+
"step": 14000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 17.994858611825194,
|
| 452 |
+
"eval_accuracy": 0.7779743325654593,
|
| 453 |
+
"eval_loss": 1.0130186080932617,
|
| 454 |
+
"eval_runtime": 32.7375,
|
| 455 |
+
"eval_samples_per_second": 190.088,
|
| 456 |
+
"eval_steps_per_second": 5.956,
|
| 457 |
+
"step": 14000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 18.637532133676093,
|
| 461 |
+
"grad_norm": 2.590914249420166,
|
| 462 |
+
"learning_rate": 8.55e-05,
|
| 463 |
+
"loss": 0.9928,
|
| 464 |
+
"step": 14500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 18.637532133676093,
|
| 468 |
+
"eval_accuracy": 0.7779691811484551,
|
| 469 |
+
"eval_loss": 1.0204987525939941,
|
| 470 |
+
"eval_runtime": 33.4597,
|
| 471 |
+
"eval_samples_per_second": 185.985,
|
| 472 |
+
"eval_steps_per_second": 5.828,
|
| 473 |
+
"step": 14500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 19.28020565552699,
|
| 477 |
+
"grad_norm": 2.6548004150390625,
|
| 478 |
+
"learning_rate": 8.5e-05,
|
| 479 |
+
"loss": 0.985,
|
| 480 |
+
"step": 15000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 19.28020565552699,
|
| 484 |
+
"eval_accuracy": 0.7788039344262295,
|
| 485 |
+
"eval_loss": 1.0153322219848633,
|
| 486 |
+
"eval_runtime": 34.4688,
|
| 487 |
+
"eval_samples_per_second": 180.54,
|
| 488 |
+
"eval_steps_per_second": 5.657,
|
| 489 |
+
"step": 15000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 19.922879177377894,
|
| 493 |
+
"grad_norm": 2.6247718334198,
|
| 494 |
+
"learning_rate": 8.450000000000001e-05,
|
| 495 |
+
"loss": 0.9822,
|
| 496 |
+
"step": 15500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 19.922879177377894,
|
| 500 |
+
"eval_accuracy": 0.7795580110497238,
|
| 501 |
+
"eval_loss": 1.0121690034866333,
|
| 502 |
+
"eval_runtime": 34.3855,
|
| 503 |
+
"eval_samples_per_second": 180.977,
|
| 504 |
+
"eval_steps_per_second": 5.671,
|
| 505 |
+
"step": 15500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 20.565552699228792,
|
| 509 |
+
"grad_norm": 2.7304604053497314,
|
| 510 |
+
"learning_rate": 8.4e-05,
|
| 511 |
+
"loss": 0.9735,
|
| 512 |
+
"step": 16000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 20.565552699228792,
|
| 516 |
+
"eval_accuracy": 0.7793682688302468,
|
| 517 |
+
"eval_loss": 1.018441081047058,
|
| 518 |
+
"eval_runtime": 34.5499,
|
| 519 |
+
"eval_samples_per_second": 180.116,
|
| 520 |
+
"eval_steps_per_second": 5.644,
|
| 521 |
+
"step": 16000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 21.20822622107969,
|
| 525 |
+
"grad_norm": 2.7982735633850098,
|
| 526 |
+
"learning_rate": 8.35e-05,
|
| 527 |
+
"loss": 0.971,
|
| 528 |
+
"step": 16500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 21.20822622107969,
|
| 532 |
+
"eval_accuracy": 0.7816012690588251,
|
| 533 |
+
"eval_loss": 1.0020060539245605,
|
| 534 |
+
"eval_runtime": 34.6648,
|
| 535 |
+
"eval_samples_per_second": 179.519,
|
| 536 |
+
"eval_steps_per_second": 5.625,
|
| 537 |
+
"step": 16500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 21.85089974293059,
|
| 541 |
+
"grad_norm": 2.6124517917633057,
|
| 542 |
+
"learning_rate": 8.3e-05,
|
| 543 |
+
"loss": 0.965,
|
| 544 |
+
"step": 17000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 21.85089974293059,
|
| 548 |
+
"eval_accuracy": 0.7828390025055424,
|
| 549 |
+
"eval_loss": 0.9919770359992981,
|
| 550 |
+
"eval_runtime": 34.2625,
|
| 551 |
+
"eval_samples_per_second": 181.627,
|
| 552 |
+
"eval_steps_per_second": 5.691,
|
| 553 |
+
"step": 17000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 22.493573264781492,
|
| 557 |
+
"grad_norm": 2.7260468006134033,
|
| 558 |
+
"learning_rate": 8.25e-05,
|
| 559 |
+
"loss": 0.9536,
|
| 560 |
+
"step": 17500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 22.493573264781492,
|
| 564 |
+
"eval_accuracy": 0.7826561078737454,
|
| 565 |
+
"eval_loss": 0.9964642524719238,
|
| 566 |
+
"eval_runtime": 35.1121,
|
| 567 |
+
"eval_samples_per_second": 177.232,
|
| 568 |
+
"eval_steps_per_second": 5.554,
|
| 569 |
+
"step": 17500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 23.13624678663239,
|
| 573 |
+
"grad_norm": 2.5959136486053467,
|
| 574 |
+
"learning_rate": 8.2e-05,
|
| 575 |
+
"loss": 0.9515,
|
| 576 |
+
"step": 18000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 23.13624678663239,
|
| 580 |
+
"eval_accuracy": 0.7842808178702281,
|
| 581 |
+
"eval_loss": 0.9843628406524658,
|
| 582 |
+
"eval_runtime": 35.1683,
|
| 583 |
+
"eval_samples_per_second": 176.949,
|
| 584 |
+
"eval_steps_per_second": 5.545,
|
| 585 |
+
"step": 18000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 23.77892030848329,
|
| 589 |
+
"grad_norm": 2.8720030784606934,
|
| 590 |
+
"learning_rate": 8.15e-05,
|
| 591 |
+
"loss": 0.9459,
|
| 592 |
+
"step": 18500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 23.77892030848329,
|
| 596 |
+
"eval_accuracy": 0.7822331893892659,
|
| 597 |
+
"eval_loss": 0.9950876235961914,
|
| 598 |
+
"eval_runtime": 35.4899,
|
| 599 |
+
"eval_samples_per_second": 175.345,
|
| 600 |
+
"eval_steps_per_second": 5.495,
|
| 601 |
+
"step": 18500
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 24.42159383033419,
|
| 605 |
+
"grad_norm": 3.0184552669525146,
|
| 606 |
+
"learning_rate": 8.1e-05,
|
| 607 |
+
"loss": 0.9374,
|
| 608 |
+
"step": 19000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 24.42159383033419,
|
| 612 |
+
"eval_accuracy": 0.7852158669501615,
|
| 613 |
+
"eval_loss": 0.9859277009963989,
|
| 614 |
+
"eval_runtime": 34.116,
|
| 615 |
+
"eval_samples_per_second": 182.407,
|
| 616 |
+
"eval_steps_per_second": 5.716,
|
| 617 |
+
"step": 19000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 25.06426735218509,
|
| 621 |
+
"grad_norm": 2.516874074935913,
|
| 622 |
+
"learning_rate": 8.05e-05,
|
| 623 |
+
"loss": 0.9318,
|
| 624 |
+
"step": 19500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 25.06426735218509,
|
| 628 |
+
"eval_accuracy": 0.7829394214826243,
|
| 629 |
+
"eval_loss": 0.9957559108734131,
|
| 630 |
+
"eval_runtime": 35.0753,
|
| 631 |
+
"eval_samples_per_second": 177.419,
|
| 632 |
+
"eval_steps_per_second": 5.559,
|
| 633 |
+
"step": 19500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 25.70694087403599,
|
| 637 |
+
"grad_norm": 2.5600101947784424,
|
| 638 |
+
"learning_rate": 8e-05,
|
| 639 |
+
"loss": 0.9311,
|
| 640 |
+
"step": 20000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 25.70694087403599,
|
| 644 |
+
"eval_accuracy": 0.786506985609304,
|
| 645 |
+
"eval_loss": 0.9760250449180603,
|
| 646 |
+
"eval_runtime": 34.6952,
|
| 647 |
+
"eval_samples_per_second": 179.362,
|
| 648 |
+
"eval_steps_per_second": 5.62,
|
| 649 |
+
"step": 20000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 26.34961439588689,
|
| 653 |
+
"grad_norm": 2.7094664573669434,
|
| 654 |
+
"learning_rate": 7.950000000000001e-05,
|
| 655 |
+
"loss": 0.913,
|
| 656 |
+
"step": 20500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 26.34961439588689,
|
| 660 |
+
"eval_accuracy": 0.78658190904128,
|
| 661 |
+
"eval_loss": 0.9806557297706604,
|
| 662 |
+
"eval_runtime": 34.3785,
|
| 663 |
+
"eval_samples_per_second": 181.014,
|
| 664 |
+
"eval_steps_per_second": 5.672,
|
| 665 |
+
"step": 20500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 26.99228791773779,
|
| 669 |
+
"grad_norm": 2.3693931102752686,
|
| 670 |
+
"learning_rate": 7.900000000000001e-05,
|
| 671 |
+
"loss": 0.9171,
|
| 672 |
+
"step": 21000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 26.99228791773779,
|
| 676 |
+
"eval_accuracy": 0.7854132867260752,
|
| 677 |
+
"eval_loss": 0.9825689196586609,
|
| 678 |
+
"eval_runtime": 34.4341,
|
| 679 |
+
"eval_samples_per_second": 180.722,
|
| 680 |
+
"eval_steps_per_second": 5.663,
|
| 681 |
+
"step": 21000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 27.63496143958869,
|
| 685 |
+
"grad_norm": 3.0254580974578857,
|
| 686 |
+
"learning_rate": 7.850000000000001e-05,
|
| 687 |
+
"loss": 0.9062,
|
| 688 |
+
"step": 21500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 27.63496143958869,
|
| 692 |
+
"eval_accuracy": 0.7879520726883571,
|
| 693 |
+
"eval_loss": 0.9732692241668701,
|
| 694 |
+
"eval_runtime": 34.5783,
|
| 695 |
+
"eval_samples_per_second": 179.968,
|
| 696 |
+
"eval_steps_per_second": 5.639,
|
| 697 |
+
"step": 21500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 28.27763496143959,
|
| 701 |
+
"grad_norm": 2.6540822982788086,
|
| 702 |
+
"learning_rate": 7.800000000000001e-05,
|
| 703 |
+
"loss": 0.9095,
|
| 704 |
+
"step": 22000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 28.27763496143959,
|
| 708 |
+
"eval_accuracy": 0.7879547877788341,
|
| 709 |
+
"eval_loss": 0.9751275777816772,
|
| 710 |
+
"eval_runtime": 33.0181,
|
| 711 |
+
"eval_samples_per_second": 188.473,
|
| 712 |
+
"eval_steps_per_second": 5.906,
|
| 713 |
+
"step": 22000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 28.920308483290487,
|
| 717 |
+
"grad_norm": 2.7978150844573975,
|
| 718 |
+
"learning_rate": 7.75e-05,
|
| 719 |
+
"loss": 0.9003,
|
| 720 |
+
"step": 22500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 28.920308483290487,
|
| 724 |
+
"eval_accuracy": 0.789157797413295,
|
| 725 |
+
"eval_loss": 0.9676902890205383,
|
| 726 |
+
"eval_runtime": 34.176,
|
| 727 |
+
"eval_samples_per_second": 182.087,
|
| 728 |
+
"eval_steps_per_second": 5.706,
|
| 729 |
+
"step": 22500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 29.56298200514139,
|
| 733 |
+
"grad_norm": 2.6367812156677246,
|
| 734 |
+
"learning_rate": 7.7e-05,
|
| 735 |
+
"loss": 0.8994,
|
| 736 |
+
"step": 23000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 29.56298200514139,
|
| 740 |
+
"eval_accuracy": 0.78875261302136,
|
| 741 |
+
"eval_loss": 0.9720832109451294,
|
| 742 |
+
"eval_runtime": 34.6435,
|
| 743 |
+
"eval_samples_per_second": 179.63,
|
| 744 |
+
"eval_steps_per_second": 5.629,
|
| 745 |
+
"step": 23000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 30.205655526992288,
|
| 749 |
+
"grad_norm": 2.8410837650299072,
|
| 750 |
+
"learning_rate": 7.65e-05,
|
| 751 |
+
"loss": 0.8937,
|
| 752 |
+
"step": 23500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 30.205655526992288,
|
| 756 |
+
"eval_accuracy": 0.7898726325583352,
|
| 757 |
+
"eval_loss": 0.9666265249252319,
|
| 758 |
+
"eval_runtime": 34.382,
|
| 759 |
+
"eval_samples_per_second": 180.996,
|
| 760 |
+
"eval_steps_per_second": 5.672,
|
| 761 |
+
"step": 23500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 30.848329048843187,
|
| 765 |
+
"grad_norm": 2.6249313354492188,
|
| 766 |
+
"learning_rate": 7.6e-05,
|
| 767 |
+
"loss": 0.8853,
|
| 768 |
+
"step": 24000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 30.848329048843187,
|
| 772 |
+
"eval_accuracy": 0.7914164731021732,
|
| 773 |
+
"eval_loss": 0.9592417478561401,
|
| 774 |
+
"eval_runtime": 34.7355,
|
| 775 |
+
"eval_samples_per_second": 179.154,
|
| 776 |
+
"eval_steps_per_second": 5.614,
|
| 777 |
+
"step": 24000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 31.491002570694086,
|
| 781 |
+
"grad_norm": 2.563669443130493,
|
| 782 |
+
"learning_rate": 7.55e-05,
|
| 783 |
+
"loss": 0.8838,
|
| 784 |
+
"step": 24500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 31.491002570694086,
|
| 788 |
+
"eval_accuracy": 0.7892845518211694,
|
| 789 |
+
"eval_loss": 0.969780683517456,
|
| 790 |
+
"eval_runtime": 34.7542,
|
| 791 |
+
"eval_samples_per_second": 179.057,
|
| 792 |
+
"eval_steps_per_second": 5.611,
|
| 793 |
+
"step": 24500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 32.13367609254499,
|
| 797 |
+
"grad_norm": 2.7462027072906494,
|
| 798 |
+
"learning_rate": 7.500000000000001e-05,
|
| 799 |
+
"loss": 0.8809,
|
| 800 |
+
"step": 25000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 32.13367609254499,
|
| 804 |
+
"eval_accuracy": 0.7920457676713335,
|
| 805 |
+
"eval_loss": 0.9529361128807068,
|
| 806 |
+
"eval_runtime": 34.3098,
|
| 807 |
+
"eval_samples_per_second": 181.377,
|
| 808 |
+
"eval_steps_per_second": 5.684,
|
| 809 |
+
"step": 25000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 32.77634961439589,
|
| 813 |
+
"grad_norm": 2.533267021179199,
|
| 814 |
+
"learning_rate": 7.450000000000001e-05,
|
| 815 |
+
"loss": 0.874,
|
| 816 |
+
"step": 25500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 32.77634961439589,
|
| 820 |
+
"eval_accuracy": 0.7919883512638669,
|
| 821 |
+
"eval_loss": 0.9449800252914429,
|
| 822 |
+
"eval_runtime": 34.7988,
|
| 823 |
+
"eval_samples_per_second": 178.828,
|
| 824 |
+
"eval_steps_per_second": 5.604,
|
| 825 |
+
"step": 25500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 33.419023136246786,
|
| 829 |
+
"grad_norm": 2.718332529067993,
|
| 830 |
+
"learning_rate": 7.4e-05,
|
| 831 |
+
"loss": 0.8647,
|
| 832 |
+
"step": 26000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 33.419023136246786,
|
| 836 |
+
"eval_accuracy": 0.7927094725330193,
|
| 837 |
+
"eval_loss": 0.9489922523498535,
|
| 838 |
+
"eval_runtime": 34.773,
|
| 839 |
+
"eval_samples_per_second": 178.961,
|
| 840 |
+
"eval_steps_per_second": 5.608,
|
| 841 |
+
"step": 26000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 34.06169665809769,
|
| 845 |
+
"grad_norm": 2.664407253265381,
|
| 846 |
+
"learning_rate": 7.35e-05,
|
| 847 |
+
"loss": 0.8638,
|
| 848 |
+
"step": 26500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 34.06169665809769,
|
| 852 |
+
"eval_accuracy": 0.7909423101119826,
|
| 853 |
+
"eval_loss": 0.9638357758522034,
|
| 854 |
+
"eval_runtime": 34.4764,
|
| 855 |
+
"eval_samples_per_second": 180.501,
|
| 856 |
+
"eval_steps_per_second": 5.656,
|
| 857 |
+
"step": 26500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 34.70437017994858,
|
| 861 |
+
"grad_norm": 2.9981689453125,
|
| 862 |
+
"learning_rate": 7.3e-05,
|
| 863 |
+
"loss": 0.8604,
|
| 864 |
+
"step": 27000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 34.70437017994858,
|
| 868 |
+
"eval_accuracy": 0.792258094990671,
|
| 869 |
+
"eval_loss": 0.9567773938179016,
|
| 870 |
+
"eval_runtime": 34.4124,
|
| 871 |
+
"eval_samples_per_second": 180.836,
|
| 872 |
+
"eval_steps_per_second": 5.667,
|
| 873 |
+
"step": 27000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 35.347043701799485,
|
| 877 |
+
"grad_norm": 2.6397457122802734,
|
| 878 |
+
"learning_rate": 7.25e-05,
|
| 879 |
+
"loss": 0.8516,
|
| 880 |
+
"step": 27500
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 35.347043701799485,
|
| 884 |
+
"eval_accuracy": 0.7944613334662306,
|
| 885 |
+
"eval_loss": 0.9480540156364441,
|
| 886 |
+
"eval_runtime": 34.3734,
|
| 887 |
+
"eval_samples_per_second": 181.041,
|
| 888 |
+
"eval_steps_per_second": 5.673,
|
| 889 |
+
"step": 27500
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 35.98971722365039,
|
| 893 |
+
"grad_norm": 2.9263813495635986,
|
| 894 |
+
"learning_rate": 7.2e-05,
|
| 895 |
+
"loss": 0.8529,
|
| 896 |
+
"step": 28000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 35.98971722365039,
|
| 900 |
+
"eval_accuracy": 0.7949814058148135,
|
| 901 |
+
"eval_loss": 0.9380202293395996,
|
| 902 |
+
"eval_runtime": 34.1807,
|
| 903 |
+
"eval_samples_per_second": 182.062,
|
| 904 |
+
"eval_steps_per_second": 5.705,
|
| 905 |
+
"step": 28000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 36.63239074550128,
|
| 909 |
+
"grad_norm": 2.446913719177246,
|
| 910 |
+
"learning_rate": 7.15e-05,
|
| 911 |
+
"loss": 0.8454,
|
| 912 |
+
"step": 28500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 36.63239074550128,
|
| 916 |
+
"eval_accuracy": 0.7940900415285876,
|
| 917 |
+
"eval_loss": 0.9452695250511169,
|
| 918 |
+
"eval_runtime": 33.1944,
|
| 919 |
+
"eval_samples_per_second": 187.472,
|
| 920 |
+
"eval_steps_per_second": 5.874,
|
| 921 |
+
"step": 28500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 37.275064267352185,
|
| 925 |
+
"grad_norm": 2.429306983947754,
|
| 926 |
+
"learning_rate": 7.1e-05,
|
| 927 |
+
"loss": 0.8432,
|
| 928 |
+
"step": 29000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 37.275064267352185,
|
| 932 |
+
"eval_accuracy": 0.7944350218702906,
|
| 933 |
+
"eval_loss": 0.9466774463653564,
|
| 934 |
+
"eval_runtime": 33.291,
|
| 935 |
+
"eval_samples_per_second": 186.927,
|
| 936 |
+
"eval_steps_per_second": 5.857,
|
| 937 |
+
"step": 29000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 37.91773778920309,
|
| 941 |
+
"grad_norm": 2.648399591445923,
|
| 942 |
+
"learning_rate": 7.05e-05,
|
| 943 |
+
"loss": 0.8468,
|
| 944 |
+
"step": 29500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 37.91773778920309,
|
| 948 |
+
"eval_accuracy": 0.7957481551099058,
|
| 949 |
+
"eval_loss": 0.9400731325149536,
|
| 950 |
+
"eval_runtime": 33.2405,
|
| 951 |
+
"eval_samples_per_second": 187.211,
|
| 952 |
+
"eval_steps_per_second": 5.866,
|
| 953 |
+
"step": 29500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 38.56041131105398,
|
| 957 |
+
"grad_norm": 2.644608974456787,
|
| 958 |
+
"learning_rate": 7e-05,
|
| 959 |
+
"loss": 0.8328,
|
| 960 |
+
"step": 30000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 38.56041131105398,
|
| 964 |
+
"eval_accuracy": 0.795926071569013,
|
| 965 |
+
"eval_loss": 0.9422577619552612,
|
| 966 |
+
"eval_runtime": 33.3493,
|
| 967 |
+
"eval_samples_per_second": 186.601,
|
| 968 |
+
"eval_steps_per_second": 5.847,
|
| 969 |
+
"step": 30000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 39.203084832904885,
|
| 973 |
+
"grad_norm": 2.6563777923583984,
|
| 974 |
+
"learning_rate": 6.95e-05,
|
| 975 |
+
"loss": 0.8323,
|
| 976 |
+
"step": 30500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 39.203084832904885,
|
| 980 |
+
"eval_accuracy": 0.7951947535355016,
|
| 981 |
+
"eval_loss": 0.9368035793304443,
|
| 982 |
+
"eval_runtime": 32.9628,
|
| 983 |
+
"eval_samples_per_second": 188.788,
|
| 984 |
+
"eval_steps_per_second": 5.916,
|
| 985 |
+
"step": 30500
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 39.84575835475579,
|
| 989 |
+
"grad_norm": 2.5932886600494385,
|
| 990 |
+
"learning_rate": 6.9e-05,
|
| 991 |
+
"loss": 0.8268,
|
| 992 |
+
"step": 31000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 39.84575835475579,
|
| 996 |
+
"eval_accuracy": 0.7967557476448579,
|
| 997 |
+
"eval_loss": 0.9303238391876221,
|
| 998 |
+
"eval_runtime": 32.4851,
|
| 999 |
+
"eval_samples_per_second": 191.565,
|
| 1000 |
+
"eval_steps_per_second": 6.003,
|
| 1001 |
+
"step": 31000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 40.48843187660668,
|
| 1005 |
+
"grad_norm": 2.7462244033813477,
|
| 1006 |
+
"learning_rate": 6.850000000000001e-05,
|
| 1007 |
+
"loss": 0.8246,
|
| 1008 |
+
"step": 31500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 40.48843187660668,
|
| 1012 |
+
"eval_accuracy": 0.7966305277950054,
|
| 1013 |
+
"eval_loss": 0.9323325157165527,
|
| 1014 |
+
"eval_runtime": 33.1165,
|
| 1015 |
+
"eval_samples_per_second": 187.912,
|
| 1016 |
+
"eval_steps_per_second": 5.888,
|
| 1017 |
+
"step": 31500
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 41.131105398457585,
|
| 1021 |
+
"grad_norm": 2.647944211959839,
|
| 1022 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1023 |
+
"loss": 0.8216,
|
| 1024 |
+
"step": 32000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 41.131105398457585,
|
| 1028 |
+
"eval_accuracy": 0.7962699125415105,
|
| 1029 |
+
"eval_loss": 0.9350172281265259,
|
| 1030 |
+
"eval_runtime": 32.873,
|
| 1031 |
+
"eval_samples_per_second": 189.305,
|
| 1032 |
+
"eval_steps_per_second": 5.932,
|
| 1033 |
+
"step": 32000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 41.77377892030848,
|
| 1037 |
+
"grad_norm": 2.661036729812622,
|
| 1038 |
+
"learning_rate": 6.750000000000001e-05,
|
| 1039 |
+
"loss": 0.8218,
|
| 1040 |
+
"step": 32500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 41.77377892030848,
|
| 1044 |
+
"eval_accuracy": 0.7972857023271388,
|
| 1045 |
+
"eval_loss": 0.9303329586982727,
|
| 1046 |
+
"eval_runtime": 32.6506,
|
| 1047 |
+
"eval_samples_per_second": 190.594,
|
| 1048 |
+
"eval_steps_per_second": 5.972,
|
| 1049 |
+
"step": 32500
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 42.41645244215938,
|
| 1053 |
+
"grad_norm": 2.7694153785705566,
|
| 1054 |
+
"learning_rate": 6.7e-05,
|
| 1055 |
+
"loss": 0.8132,
|
| 1056 |
+
"step": 33000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 42.41645244215938,
|
| 1060 |
+
"eval_accuracy": 0.7977203787112787,
|
| 1061 |
+
"eval_loss": 0.9363131523132324,
|
| 1062 |
+
"eval_runtime": 33.0889,
|
| 1063 |
+
"eval_samples_per_second": 188.069,
|
| 1064 |
+
"eval_steps_per_second": 5.893,
|
| 1065 |
+
"step": 33000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 43.059125964010285,
|
| 1069 |
+
"grad_norm": 2.8761026859283447,
|
| 1070 |
+
"learning_rate": 6.65e-05,
|
| 1071 |
+
"loss": 0.8082,
|
| 1072 |
+
"step": 33500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 43.059125964010285,
|
| 1076 |
+
"eval_accuracy": 0.7981091852326369,
|
| 1077 |
+
"eval_loss": 0.9364966154098511,
|
| 1078 |
+
"eval_runtime": 32.3923,
|
| 1079 |
+
"eval_samples_per_second": 192.113,
|
| 1080 |
+
"eval_steps_per_second": 6.02,
|
| 1081 |
+
"step": 33500
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 43.70179948586118,
|
| 1085 |
+
"grad_norm": 2.6908044815063477,
|
| 1086 |
+
"learning_rate": 6.6e-05,
|
| 1087 |
+
"loss": 0.806,
|
| 1088 |
+
"step": 34000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 43.70179948586118,
|
| 1092 |
+
"eval_accuracy": 0.7986442847736496,
|
| 1093 |
+
"eval_loss": 0.9224198460578918,
|
| 1094 |
+
"eval_runtime": 33.388,
|
| 1095 |
+
"eval_samples_per_second": 186.384,
|
| 1096 |
+
"eval_steps_per_second": 5.84,
|
| 1097 |
+
"step": 34000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 44.34447300771208,
|
| 1101 |
+
"grad_norm": 2.6623470783233643,
|
| 1102 |
+
"learning_rate": 6.55e-05,
|
| 1103 |
+
"loss": 0.8024,
|
| 1104 |
+
"step": 34500
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 44.34447300771208,
|
| 1108 |
+
"eval_accuracy": 0.7992224783044672,
|
| 1109 |
+
"eval_loss": 0.9192689061164856,
|
| 1110 |
+
"eval_runtime": 33.3722,
|
| 1111 |
+
"eval_samples_per_second": 186.473,
|
| 1112 |
+
"eval_steps_per_second": 5.843,
|
| 1113 |
+
"step": 34500
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 44.987146529562985,
|
| 1117 |
+
"grad_norm": 2.6698567867279053,
|
| 1118 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1119 |
+
"loss": 0.8016,
|
| 1120 |
+
"step": 35000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 44.987146529562985,
|
| 1124 |
+
"eval_accuracy": 0.7984851112126992,
|
| 1125 |
+
"eval_loss": 0.9178963303565979,
|
| 1126 |
+
"eval_runtime": 33.2663,
|
| 1127 |
+
"eval_samples_per_second": 187.066,
|
| 1128 |
+
"eval_steps_per_second": 5.862,
|
| 1129 |
+
"step": 35000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 45.62982005141388,
|
| 1133 |
+
"grad_norm": 2.980325222015381,
|
| 1134 |
+
"learning_rate": 6.450000000000001e-05,
|
| 1135 |
+
"loss": 0.796,
|
| 1136 |
+
"step": 35500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 45.62982005141388,
|
| 1140 |
+
"eval_accuracy": 0.7979037090039865,
|
| 1141 |
+
"eval_loss": 0.9273726344108582,
|
| 1142 |
+
"eval_runtime": 33.0665,
|
| 1143 |
+
"eval_samples_per_second": 188.196,
|
| 1144 |
+
"eval_steps_per_second": 5.897,
|
| 1145 |
+
"step": 35500
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 46.27249357326478,
|
| 1149 |
+
"grad_norm": 2.536480188369751,
|
| 1150 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1151 |
+
"loss": 0.7976,
|
| 1152 |
+
"step": 36000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 46.27249357326478,
|
| 1156 |
+
"eval_accuracy": 0.8009755918651776,
|
| 1157 |
+
"eval_loss": 0.9136722087860107,
|
| 1158 |
+
"eval_runtime": 33.7816,
|
| 1159 |
+
"eval_samples_per_second": 184.213,
|
| 1160 |
+
"eval_steps_per_second": 5.772,
|
| 1161 |
+
"step": 36000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 46.915167095115685,
|
| 1165 |
+
"grad_norm": 2.369147777557373,
|
| 1166 |
+
"learning_rate": 6.35e-05,
|
| 1167 |
+
"loss": 0.7888,
|
| 1168 |
+
"step": 36500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 46.915167095115685,
|
| 1172 |
+
"eval_accuracy": 0.8006721188810271,
|
| 1173 |
+
"eval_loss": 0.9139747619628906,
|
| 1174 |
+
"eval_runtime": 33.0443,
|
| 1175 |
+
"eval_samples_per_second": 188.323,
|
| 1176 |
+
"eval_steps_per_second": 5.901,
|
| 1177 |
+
"step": 36500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 47.55784061696658,
|
| 1181 |
+
"grad_norm": 2.606424570083618,
|
| 1182 |
+
"learning_rate": 6.3e-05,
|
| 1183 |
+
"loss": 0.7826,
|
| 1184 |
+
"step": 37000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 47.55784061696658,
|
| 1188 |
+
"eval_accuracy": 0.8006546893234877,
|
| 1189 |
+
"eval_loss": 0.9165197610855103,
|
| 1190 |
+
"eval_runtime": 32.9373,
|
| 1191 |
+
"eval_samples_per_second": 188.935,
|
| 1192 |
+
"eval_steps_per_second": 5.92,
|
| 1193 |
+
"step": 37000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 48.20051413881748,
|
| 1197 |
+
"grad_norm": 2.869199514389038,
|
| 1198 |
+
"learning_rate": 6.25e-05,
|
| 1199 |
+
"loss": 0.789,
|
| 1200 |
+
"step": 37500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 48.20051413881748,
|
| 1204 |
+
"eval_accuracy": 0.8009030033534118,
|
| 1205 |
+
"eval_loss": 0.9167375564575195,
|
| 1206 |
+
"eval_runtime": 33.2315,
|
| 1207 |
+
"eval_samples_per_second": 187.262,
|
| 1208 |
+
"eval_steps_per_second": 5.868,
|
| 1209 |
+
"step": 37500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 48.84318766066838,
|
| 1213 |
+
"grad_norm": 2.5320894718170166,
|
| 1214 |
+
"learning_rate": 6.2e-05,
|
| 1215 |
+
"loss": 0.7827,
|
| 1216 |
+
"step": 38000
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 48.84318766066838,
|
| 1220 |
+
"eval_accuracy": 0.7997841239160237,
|
| 1221 |
+
"eval_loss": 0.9282008409500122,
|
| 1222 |
+
"eval_runtime": 33.3582,
|
| 1223 |
+
"eval_samples_per_second": 186.551,
|
| 1224 |
+
"eval_steps_per_second": 5.846,
|
| 1225 |
+
"step": 38000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 49.48586118251928,
|
| 1229 |
+
"grad_norm": 2.741992712020874,
|
| 1230 |
+
"learning_rate": 6.15e-05,
|
| 1231 |
+
"loss": 0.7739,
|
| 1232 |
+
"step": 38500
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 49.48586118251928,
|
| 1236 |
+
"eval_accuracy": 0.8020669565628894,
|
| 1237 |
+
"eval_loss": 0.9078426361083984,
|
| 1238 |
+
"eval_runtime": 33.2299,
|
| 1239 |
+
"eval_samples_per_second": 187.271,
|
| 1240 |
+
"eval_steps_per_second": 5.868,
|
| 1241 |
+
"step": 38500
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 50.12853470437018,
|
| 1245 |
+
"grad_norm": 2.563089370727539,
|
| 1246 |
+
"learning_rate": 6.1e-05,
|
| 1247 |
+
"loss": 0.7766,
|
| 1248 |
+
"step": 39000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 50.12853470437018,
|
| 1252 |
+
"eval_accuracy": 0.8028247073903633,
|
| 1253 |
+
"eval_loss": 0.9107823967933655,
|
| 1254 |
+
"eval_runtime": 33.5047,
|
| 1255 |
+
"eval_samples_per_second": 185.735,
|
| 1256 |
+
"eval_steps_per_second": 5.82,
|
| 1257 |
+
"step": 39000
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 50.77120822622108,
|
| 1261 |
+
"grad_norm": 2.5732595920562744,
|
| 1262 |
+
"learning_rate": 6.05e-05,
|
| 1263 |
+
"loss": 0.7728,
|
| 1264 |
+
"step": 39500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 50.77120822622108,
|
| 1268 |
+
"eval_accuracy": 0.8025567474818919,
|
| 1269 |
+
"eval_loss": 0.9065195918083191,
|
| 1270 |
+
"eval_runtime": 33.1402,
|
| 1271 |
+
"eval_samples_per_second": 187.778,
|
| 1272 |
+
"eval_steps_per_second": 5.884,
|
| 1273 |
+
"step": 39500
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 51.41388174807198,
|
| 1277 |
+
"grad_norm": 2.4717209339141846,
|
| 1278 |
+
"learning_rate": 6e-05,
|
| 1279 |
+
"loss": 0.7676,
|
| 1280 |
+
"step": 40000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 51.41388174807198,
|
| 1284 |
+
"eval_accuracy": 0.8028714592601541,
|
| 1285 |
+
"eval_loss": 0.9060749411582947,
|
| 1286 |
+
"eval_runtime": 32.3856,
|
| 1287 |
+
"eval_samples_per_second": 192.153,
|
| 1288 |
+
"eval_steps_per_second": 6.021,
|
| 1289 |
+
"step": 40000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 52.05655526992288,
|
| 1293 |
+
"grad_norm": 2.5366320610046387,
|
| 1294 |
+
"learning_rate": 5.95e-05,
|
| 1295 |
+
"loss": 0.7653,
|
| 1296 |
+
"step": 40500
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 52.05655526992288,
|
| 1300 |
+
"eval_accuracy": 0.8030310980056771,
|
| 1301 |
+
"eval_loss": 0.9148956537246704,
|
| 1302 |
+
"eval_runtime": 33.3546,
|
| 1303 |
+
"eval_samples_per_second": 186.571,
|
| 1304 |
+
"eval_steps_per_second": 5.846,
|
| 1305 |
+
"step": 40500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 52.69922879177378,
|
| 1309 |
+
"grad_norm": 2.9074325561523438,
|
| 1310 |
+
"learning_rate": 5.9e-05,
|
| 1311 |
+
"loss": 0.7607,
|
| 1312 |
+
"step": 41000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 52.69922879177378,
|
| 1316 |
+
"eval_accuracy": 0.8039304797381431,
|
| 1317 |
+
"eval_loss": 0.8974832892417908,
|
| 1318 |
+
"eval_runtime": 33.4483,
|
| 1319 |
+
"eval_samples_per_second": 186.049,
|
| 1320 |
+
"eval_steps_per_second": 5.83,
|
| 1321 |
+
"step": 41000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 53.34190231362468,
|
| 1325 |
+
"grad_norm": 3.1211259365081787,
|
| 1326 |
+
"learning_rate": 5.85e-05,
|
| 1327 |
+
"loss": 0.7604,
|
| 1328 |
+
"step": 41500
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 53.34190231362468,
|
| 1332 |
+
"eval_accuracy": 0.8036355728594897,
|
| 1333 |
+
"eval_loss": 0.90870201587677,
|
| 1334 |
+
"eval_runtime": 33.4646,
|
| 1335 |
+
"eval_samples_per_second": 185.957,
|
| 1336 |
+
"eval_steps_per_second": 5.827,
|
| 1337 |
+
"step": 41500
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 53.98457583547558,
|
| 1341 |
+
"grad_norm": 2.762848377227783,
|
| 1342 |
+
"learning_rate": 5.8e-05,
|
| 1343 |
+
"loss": 0.7593,
|
| 1344 |
+
"step": 42000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 53.98457583547558,
|
| 1348 |
+
"eval_accuracy": 0.802461854621613,
|
| 1349 |
+
"eval_loss": 0.9101512432098389,
|
| 1350 |
+
"eval_runtime": 32.3612,
|
| 1351 |
+
"eval_samples_per_second": 192.298,
|
| 1352 |
+
"eval_steps_per_second": 6.026,
|
| 1353 |
+
"step": 42000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 54.62724935732648,
|
| 1357 |
+
"grad_norm": 2.6548171043395996,
|
| 1358 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 1359 |
+
"loss": 0.7564,
|
| 1360 |
+
"step": 42500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 54.62724935732648,
|
| 1364 |
+
"eval_accuracy": 0.804102518442478,
|
| 1365 |
+
"eval_loss": 0.9050089120864868,
|
| 1366 |
+
"eval_runtime": 33.1101,
|
| 1367 |
+
"eval_samples_per_second": 187.949,
|
| 1368 |
+
"eval_steps_per_second": 5.889,
|
| 1369 |
+
"step": 42500
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 55.26992287917738,
|
| 1373 |
+
"grad_norm": 2.6228950023651123,
|
| 1374 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1375 |
+
"loss": 0.7452,
|
| 1376 |
+
"step": 43000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 55.26992287917738,
|
| 1380 |
+
"eval_accuracy": 0.8040075271997299,
|
| 1381 |
+
"eval_loss": 0.9055400490760803,
|
| 1382 |
+
"eval_runtime": 33.014,
|
| 1383 |
+
"eval_samples_per_second": 188.496,
|
| 1384 |
+
"eval_steps_per_second": 5.907,
|
| 1385 |
+
"step": 43000
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 55.912596401028274,
|
| 1389 |
+
"grad_norm": 2.6490426063537598,
|
| 1390 |
+
"learning_rate": 5.65e-05,
|
| 1391 |
+
"loss": 0.7488,
|
| 1392 |
+
"step": 43500
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 55.912596401028274,
|
| 1396 |
+
"eval_accuracy": 0.8048906062428912,
|
| 1397 |
+
"eval_loss": 0.9053667187690735,
|
| 1398 |
+
"eval_runtime": 32.3856,
|
| 1399 |
+
"eval_samples_per_second": 192.153,
|
| 1400 |
+
"eval_steps_per_second": 6.021,
|
| 1401 |
+
"step": 43500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 56.55526992287918,
|
| 1405 |
+
"grad_norm": 2.494063377380371,
|
| 1406 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1407 |
+
"loss": 0.7472,
|
| 1408 |
+
"step": 44000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 56.55526992287918,
|
| 1412 |
+
"eval_accuracy": 0.8050882383051106,
|
| 1413 |
+
"eval_loss": 0.8990674614906311,
|
| 1414 |
+
"eval_runtime": 34.0904,
|
| 1415 |
+
"eval_samples_per_second": 182.544,
|
| 1416 |
+
"eval_steps_per_second": 5.72,
|
| 1417 |
+
"step": 44000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 57.19794344473008,
|
| 1421 |
+
"grad_norm": 2.8385369777679443,
|
| 1422 |
+
"learning_rate": 5.550000000000001e-05,
|
| 1423 |
+
"loss": 0.7469,
|
| 1424 |
+
"step": 44500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 57.19794344473008,
|
| 1428 |
+
"eval_accuracy": 0.8047514040498369,
|
| 1429 |
+
"eval_loss": 0.9029610753059387,
|
| 1430 |
+
"eval_runtime": 32.596,
|
| 1431 |
+
"eval_samples_per_second": 190.913,
|
| 1432 |
+
"eval_steps_per_second": 5.982,
|
| 1433 |
+
"step": 44500
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 57.840616966580974,
|
| 1437 |
+
"grad_norm": 2.4786102771759033,
|
| 1438 |
+
"learning_rate": 5.500000000000001e-05,
|
| 1439 |
+
"loss": 0.7386,
|
| 1440 |
+
"step": 45000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 57.840616966580974,
|
| 1444 |
+
"eval_accuracy": 0.8045387474256496,
|
| 1445 |
+
"eval_loss": 0.8992837071418762,
|
| 1446 |
+
"eval_runtime": 33.7728,
|
| 1447 |
+
"eval_samples_per_second": 184.261,
|
| 1448 |
+
"eval_steps_per_second": 5.774,
|
| 1449 |
+
"step": 45000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 58.48329048843188,
|
| 1453 |
+
"grad_norm": 2.57100248336792,
|
| 1454 |
+
"learning_rate": 5.45e-05,
|
| 1455 |
+
"loss": 0.7383,
|
| 1456 |
+
"step": 45500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 58.48329048843188,
|
| 1460 |
+
"eval_accuracy": 0.8045938243522368,
|
| 1461 |
+
"eval_loss": 0.9126896858215332,
|
| 1462 |
+
"eval_runtime": 33.0001,
|
| 1463 |
+
"eval_samples_per_second": 188.575,
|
| 1464 |
+
"eval_steps_per_second": 5.909,
|
| 1465 |
+
"step": 45500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 59.12596401028278,
|
| 1469 |
+
"grad_norm": 2.6267364025115967,
|
| 1470 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1471 |
+
"loss": 0.7372,
|
| 1472 |
+
"step": 46000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 59.12596401028278,
|
| 1476 |
+
"eval_accuracy": 0.8047429982319032,
|
| 1477 |
+
"eval_loss": 0.9037800431251526,
|
| 1478 |
+
"eval_runtime": 33.4956,
|
| 1479 |
+
"eval_samples_per_second": 185.786,
|
| 1480 |
+
"eval_steps_per_second": 5.822,
|
| 1481 |
+
"step": 46000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 59.768637532133674,
|
| 1485 |
+
"grad_norm": 2.6067042350769043,
|
| 1486 |
+
"learning_rate": 5.3500000000000006e-05,
|
| 1487 |
+
"loss": 0.7293,
|
| 1488 |
+
"step": 46500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 59.768637532133674,
|
| 1492 |
+
"eval_accuracy": 0.8063738822187712,
|
| 1493 |
+
"eval_loss": 0.891742467880249,
|
| 1494 |
+
"eval_runtime": 33.3465,
|
| 1495 |
+
"eval_samples_per_second": 186.616,
|
| 1496 |
+
"eval_steps_per_second": 5.848,
|
| 1497 |
+
"step": 46500
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 60.411311053984576,
|
| 1501 |
+
"grad_norm": 2.748093843460083,
|
| 1502 |
+
"learning_rate": 5.300000000000001e-05,
|
| 1503 |
+
"loss": 0.7266,
|
| 1504 |
+
"step": 47000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 60.411311053984576,
|
| 1508 |
+
"eval_accuracy": 0.8060851226391891,
|
| 1509 |
+
"eval_loss": 0.8950145840644836,
|
| 1510 |
+
"eval_runtime": 33.417,
|
| 1511 |
+
"eval_samples_per_second": 186.223,
|
| 1512 |
+
"eval_steps_per_second": 5.835,
|
| 1513 |
+
"step": 47000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 61.05398457583548,
|
| 1517 |
+
"grad_norm": 2.7752370834350586,
|
| 1518 |
+
"learning_rate": 5.25e-05,
|
| 1519 |
+
"loss": 0.7278,
|
| 1520 |
+
"step": 47500
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 61.05398457583548,
|
| 1524 |
+
"eval_accuracy": 0.8052570807938081,
|
| 1525 |
+
"eval_loss": 0.9001266956329346,
|
| 1526 |
+
"eval_runtime": 33.2575,
|
| 1527 |
+
"eval_samples_per_second": 187.116,
|
| 1528 |
+
"eval_steps_per_second": 5.863,
|
| 1529 |
+
"step": 47500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 61.696658097686374,
|
| 1533 |
+
"grad_norm": 2.4993159770965576,
|
| 1534 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1535 |
+
"loss": 0.7224,
|
| 1536 |
+
"step": 48000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 61.696658097686374,
|
| 1540 |
+
"eval_accuracy": 0.8063545774342619,
|
| 1541 |
+
"eval_loss": 0.9000985622406006,
|
| 1542 |
+
"eval_runtime": 32.5061,
|
| 1543 |
+
"eval_samples_per_second": 191.441,
|
| 1544 |
+
"eval_steps_per_second": 5.999,
|
| 1545 |
+
"step": 48000
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 62.339331619537276,
|
| 1549 |
+
"grad_norm": 2.795344829559326,
|
| 1550 |
+
"learning_rate": 5.1500000000000005e-05,
|
| 1551 |
+
"loss": 0.7156,
|
| 1552 |
+
"step": 48500
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 62.339331619537276,
|
| 1556 |
+
"eval_accuracy": 0.8078125860121053,
|
| 1557 |
+
"eval_loss": 0.8850185871124268,
|
| 1558 |
+
"eval_runtime": 33.3065,
|
| 1559 |
+
"eval_samples_per_second": 186.84,
|
| 1560 |
+
"eval_steps_per_second": 5.855,
|
| 1561 |
+
"step": 48500
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 62.98200514138817,
|
| 1565 |
+
"grad_norm": 2.4736196994781494,
|
| 1566 |
+
"learning_rate": 5.1000000000000006e-05,
|
| 1567 |
+
"loss": 0.7179,
|
| 1568 |
+
"step": 49000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 62.98200514138817,
|
| 1572 |
+
"eval_accuracy": 0.8069766099061457,
|
| 1573 |
+
"eval_loss": 0.8914857506752014,
|
| 1574 |
+
"eval_runtime": 33.2699,
|
| 1575 |
+
"eval_samples_per_second": 187.046,
|
| 1576 |
+
"eval_steps_per_second": 5.861,
|
| 1577 |
+
"step": 49000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 63.624678663239074,
|
| 1581 |
+
"grad_norm": 2.7739675045013428,
|
| 1582 |
+
"learning_rate": 5.05e-05,
|
| 1583 |
+
"loss": 0.7158,
|
| 1584 |
+
"step": 49500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 63.624678663239074,
|
| 1588 |
+
"eval_accuracy": 0.806712037699331,
|
| 1589 |
+
"eval_loss": 0.8928523063659668,
|
| 1590 |
+
"eval_runtime": 33.5382,
|
| 1591 |
+
"eval_samples_per_second": 185.55,
|
| 1592 |
+
"eval_steps_per_second": 5.814,
|
| 1593 |
+
"step": 49500
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 64.26735218508998,
|
| 1597 |
+
"grad_norm": 2.6932947635650635,
|
| 1598 |
+
"learning_rate": 5e-05,
|
| 1599 |
+
"loss": 0.711,
|
| 1600 |
+
"step": 50000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 64.26735218508998,
|
| 1604 |
+
"eval_accuracy": 0.8060159735552406,
|
| 1605 |
+
"eval_loss": 0.9013872742652893,
|
| 1606 |
+
"eval_runtime": 33.9301,
|
| 1607 |
+
"eval_samples_per_second": 183.407,
|
| 1608 |
+
"eval_steps_per_second": 5.747,
|
| 1609 |
+
"step": 50000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 64.91002570694087,
|
| 1613 |
+
"grad_norm": 2.7735486030578613,
|
| 1614 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 1615 |
+
"loss": 0.7097,
|
| 1616 |
+
"step": 50500
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 64.91002570694087,
|
| 1620 |
+
"eval_accuracy": 0.8067475980060699,
|
| 1621 |
+
"eval_loss": 0.8957981467247009,
|
| 1622 |
+
"eval_runtime": 32.2026,
|
| 1623 |
+
"eval_samples_per_second": 193.245,
|
| 1624 |
+
"eval_steps_per_second": 6.055,
|
| 1625 |
+
"step": 50500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 65.55269922879178,
|
| 1629 |
+
"grad_norm": 2.9156525135040283,
|
| 1630 |
+
"learning_rate": 4.9e-05,
|
| 1631 |
+
"loss": 0.7079,
|
| 1632 |
+
"step": 51000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 65.55269922879178,
|
| 1636 |
+
"eval_accuracy": 0.8081258363288117,
|
| 1637 |
+
"eval_loss": 0.894716739654541,
|
| 1638 |
+
"eval_runtime": 32.7636,
|
| 1639 |
+
"eval_samples_per_second": 189.936,
|
| 1640 |
+
"eval_steps_per_second": 5.952,
|
| 1641 |
+
"step": 51000
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 66.19537275064268,
|
| 1645 |
+
"grad_norm": 2.6359803676605225,
|
| 1646 |
+
"learning_rate": 4.85e-05,
|
| 1647 |
+
"loss": 0.7091,
|
| 1648 |
+
"step": 51500
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 66.19537275064268,
|
| 1652 |
+
"eval_accuracy": 0.8079719372684523,
|
| 1653 |
+
"eval_loss": 0.8968275189399719,
|
| 1654 |
+
"eval_runtime": 32.9482,
|
| 1655 |
+
"eval_samples_per_second": 188.872,
|
| 1656 |
+
"eval_steps_per_second": 5.918,
|
| 1657 |
+
"step": 51500
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 66.83804627249357,
|
| 1661 |
+
"grad_norm": 2.481441020965576,
|
| 1662 |
+
"learning_rate": 4.8e-05,
|
| 1663 |
+
"loss": 0.7023,
|
| 1664 |
+
"step": 52000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 66.83804627249357,
|
| 1668 |
+
"eval_accuracy": 0.8089100382369007,
|
| 1669 |
+
"eval_loss": 0.8855522274971008,
|
| 1670 |
+
"eval_runtime": 33.2751,
|
| 1671 |
+
"eval_samples_per_second": 187.017,
|
| 1672 |
+
"eval_steps_per_second": 5.86,
|
| 1673 |
+
"step": 52000
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 67.48071979434447,
|
| 1677 |
+
"grad_norm": 2.7268731594085693,
|
| 1678 |
+
"learning_rate": 4.75e-05,
|
| 1679 |
+
"loss": 0.701,
|
| 1680 |
+
"step": 52500
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 67.48071979434447,
|
| 1684 |
+
"eval_accuracy": 0.8084944735097863,
|
| 1685 |
+
"eval_loss": 0.8864369988441467,
|
| 1686 |
+
"eval_runtime": 33.1906,
|
| 1687 |
+
"eval_samples_per_second": 187.493,
|
| 1688 |
+
"eval_steps_per_second": 5.875,
|
| 1689 |
+
"step": 52500
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 68.12339331619538,
|
| 1693 |
+
"grad_norm": 2.5931150913238525,
|
| 1694 |
+
"learning_rate": 4.7e-05,
|
| 1695 |
+
"loss": 0.6988,
|
| 1696 |
+
"step": 53000
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 68.12339331619538,
|
| 1700 |
+
"eval_accuracy": 0.8089927462608983,
|
| 1701 |
+
"eval_loss": 0.8866479992866516,
|
| 1702 |
+
"eval_runtime": 33.2058,
|
| 1703 |
+
"eval_samples_per_second": 187.407,
|
| 1704 |
+
"eval_steps_per_second": 5.872,
|
| 1705 |
+
"step": 53000
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 68.76606683804627,
|
| 1709 |
+
"grad_norm": 2.736328601837158,
|
| 1710 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 1711 |
+
"loss": 0.6994,
|
| 1712 |
+
"step": 53500
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 68.76606683804627,
|
| 1716 |
+
"eval_accuracy": 0.8084927024019897,
|
| 1717 |
+
"eval_loss": 0.8820142149925232,
|
| 1718 |
+
"eval_runtime": 32.9508,
|
| 1719 |
+
"eval_samples_per_second": 188.857,
|
| 1720 |
+
"eval_steps_per_second": 5.918,
|
| 1721 |
+
"step": 53500
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 69.40874035989717,
|
| 1725 |
+
"grad_norm": 2.8003087043762207,
|
| 1726 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1727 |
+
"loss": 0.6831,
|
| 1728 |
+
"step": 54000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 69.40874035989717,
|
| 1732 |
+
"eval_accuracy": 0.8091836520937638,
|
| 1733 |
+
"eval_loss": 0.8856648206710815,
|
| 1734 |
+
"eval_runtime": 33.3733,
|
| 1735 |
+
"eval_samples_per_second": 186.466,
|
| 1736 |
+
"eval_steps_per_second": 5.843,
|
| 1737 |
+
"step": 54000
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 70.05141388174808,
|
| 1741 |
+
"grad_norm": 2.7809290885925293,
|
| 1742 |
+
"learning_rate": 4.55e-05,
|
| 1743 |
+
"loss": 0.6948,
|
| 1744 |
+
"step": 54500
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 70.05141388174808,
|
| 1748 |
+
"eval_accuracy": 0.8089959519265054,
|
| 1749 |
+
"eval_loss": 0.89335036277771,
|
| 1750 |
+
"eval_runtime": 33.324,
|
| 1751 |
+
"eval_samples_per_second": 186.743,
|
| 1752 |
+
"eval_steps_per_second": 5.852,
|
| 1753 |
+
"step": 54500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 70.69408740359897,
|
| 1757 |
+
"grad_norm": 2.5319223403930664,
|
| 1758 |
+
"learning_rate": 4.5e-05,
|
| 1759 |
+
"loss": 0.688,
|
| 1760 |
+
"step": 55000
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 70.69408740359897,
|
| 1764 |
+
"eval_accuracy": 0.8102897851773414,
|
| 1765 |
+
"eval_loss": 0.8859269022941589,
|
| 1766 |
+
"eval_runtime": 33.0875,
|
| 1767 |
+
"eval_samples_per_second": 188.077,
|
| 1768 |
+
"eval_steps_per_second": 5.893,
|
| 1769 |
+
"step": 55000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 71.33676092544987,
|
| 1773 |
+
"grad_norm": 3.0281503200531006,
|
| 1774 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 1775 |
+
"loss": 0.6894,
|
| 1776 |
+
"step": 55500
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 71.33676092544987,
|
| 1780 |
+
"eval_accuracy": 0.8097594530479655,
|
| 1781 |
+
"eval_loss": 0.8875166773796082,
|
| 1782 |
+
"eval_runtime": 32.7248,
|
| 1783 |
+
"eval_samples_per_second": 190.161,
|
| 1784 |
+
"eval_steps_per_second": 5.959,
|
| 1785 |
+
"step": 55500
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 71.97943444730078,
|
| 1789 |
+
"grad_norm": 2.476452589035034,
|
| 1790 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1791 |
+
"loss": 0.6855,
|
| 1792 |
+
"step": 56000
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 71.97943444730078,
|
| 1796 |
+
"eval_accuracy": 0.8097316597458161,
|
| 1797 |
+
"eval_loss": 0.8884576559066772,
|
| 1798 |
+
"eval_runtime": 33.1154,
|
| 1799 |
+
"eval_samples_per_second": 187.919,
|
| 1800 |
+
"eval_steps_per_second": 5.889,
|
| 1801 |
+
"step": 56000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 72.62210796915167,
|
| 1805 |
+
"grad_norm": 2.662523031234741,
|
| 1806 |
+
"learning_rate": 4.35e-05,
|
| 1807 |
+
"loss": 0.6811,
|
| 1808 |
+
"step": 56500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 72.62210796915167,
|
| 1812 |
+
"eval_accuracy": 0.809297297864581,
|
| 1813 |
+
"eval_loss": 0.8833766579627991,
|
| 1814 |
+
"eval_runtime": 33.1424,
|
| 1815 |
+
"eval_samples_per_second": 187.766,
|
| 1816 |
+
"eval_steps_per_second": 5.884,
|
| 1817 |
+
"step": 56500
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 73.26478149100257,
|
| 1821 |
+
"grad_norm": 2.4902071952819824,
|
| 1822 |
+
"learning_rate": 4.3e-05,
|
| 1823 |
+
"loss": 0.6782,
|
| 1824 |
+
"step": 57000
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 73.26478149100257,
|
| 1828 |
+
"eval_accuracy": 0.8089903844893186,
|
| 1829 |
+
"eval_loss": 0.8910095691680908,
|
| 1830 |
+
"eval_runtime": 33.3324,
|
| 1831 |
+
"eval_samples_per_second": 186.695,
|
| 1832 |
+
"eval_steps_per_second": 5.85,
|
| 1833 |
+
"step": 57000
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 73.90745501285348,
|
| 1837 |
+
"grad_norm": 2.897806167602539,
|
| 1838 |
+
"learning_rate": 4.25e-05,
|
| 1839 |
+
"loss": 0.6761,
|
| 1840 |
+
"step": 57500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 73.90745501285348,
|
| 1844 |
+
"eval_accuracy": 0.8115941649425137,
|
| 1845 |
+
"eval_loss": 0.8758607506752014,
|
| 1846 |
+
"eval_runtime": 33.305,
|
| 1847 |
+
"eval_samples_per_second": 186.849,
|
| 1848 |
+
"eval_steps_per_second": 5.855,
|
| 1849 |
+
"step": 57500
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 74.55012853470437,
|
| 1853 |
+
"grad_norm": 2.6833012104034424,
|
| 1854 |
+
"learning_rate": 4.2e-05,
|
| 1855 |
+
"loss": 0.6713,
|
| 1856 |
+
"step": 58000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 74.55012853470437,
|
| 1860 |
+
"eval_accuracy": 0.8105924384594244,
|
| 1861 |
+
"eval_loss": 0.8842668533325195,
|
| 1862 |
+
"eval_runtime": 32.104,
|
| 1863 |
+
"eval_samples_per_second": 193.839,
|
| 1864 |
+
"eval_steps_per_second": 6.074,
|
| 1865 |
+
"step": 58000
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 75.19280205655527,
|
| 1869 |
+
"grad_norm": 3.058183193206787,
|
| 1870 |
+
"learning_rate": 4.15e-05,
|
| 1871 |
+
"loss": 0.6751,
|
| 1872 |
+
"step": 58500
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 75.19280205655527,
|
| 1876 |
+
"eval_accuracy": 0.8100959331379869,
|
| 1877 |
+
"eval_loss": 0.8922275304794312,
|
| 1878 |
+
"eval_runtime": 32.0158,
|
| 1879 |
+
"eval_samples_per_second": 194.373,
|
| 1880 |
+
"eval_steps_per_second": 6.091,
|
| 1881 |
+
"step": 58500
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 75.83547557840618,
|
| 1885 |
+
"grad_norm": 2.5572476387023926,
|
| 1886 |
+
"learning_rate": 4.1e-05,
|
| 1887 |
+
"loss": 0.667,
|
| 1888 |
+
"step": 59000
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 75.83547557840618,
|
| 1892 |
+
"eval_accuracy": 0.8115087833220307,
|
| 1893 |
+
"eval_loss": 0.8827089071273804,
|
| 1894 |
+
"eval_runtime": 32.068,
|
| 1895 |
+
"eval_samples_per_second": 194.056,
|
| 1896 |
+
"eval_steps_per_second": 6.081,
|
| 1897 |
+
"step": 59000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 76.47814910025707,
|
| 1901 |
+
"grad_norm": 2.6060643196105957,
|
| 1902 |
+
"learning_rate": 4.05e-05,
|
| 1903 |
+
"loss": 0.6628,
|
| 1904 |
+
"step": 59500
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 76.47814910025707,
|
| 1908 |
+
"eval_accuracy": 0.8114595604343708,
|
| 1909 |
+
"eval_loss": 0.8741211891174316,
|
| 1910 |
+
"eval_runtime": 32.0679,
|
| 1911 |
+
"eval_samples_per_second": 194.057,
|
| 1912 |
+
"eval_steps_per_second": 6.081,
|
| 1913 |
+
"step": 59500
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 77.12082262210797,
|
| 1917 |
+
"grad_norm": 2.643742322921753,
|
| 1918 |
+
"learning_rate": 4e-05,
|
| 1919 |
+
"loss": 0.6685,
|
| 1920 |
+
"step": 60000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 77.12082262210797,
|
| 1924 |
+
"eval_accuracy": 0.8120777359560294,
|
| 1925 |
+
"eval_loss": 0.8767244219779968,
|
| 1926 |
+
"eval_runtime": 32.012,
|
| 1927 |
+
"eval_samples_per_second": 194.396,
|
| 1928 |
+
"eval_steps_per_second": 6.091,
|
| 1929 |
+
"step": 60000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 77.76349614395887,
|
| 1933 |
+
"grad_norm": 3.0320351123809814,
|
| 1934 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 1935 |
+
"loss": 0.6618,
|
| 1936 |
+
"step": 60500
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 77.76349614395887,
|
| 1940 |
+
"eval_accuracy": 0.8108977044235925,
|
| 1941 |
+
"eval_loss": 0.882882833480835,
|
| 1942 |
+
"eval_runtime": 32.0023,
|
| 1943 |
+
"eval_samples_per_second": 194.455,
|
| 1944 |
+
"eval_steps_per_second": 6.093,
|
| 1945 |
+
"step": 60500
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 78.40616966580977,
|
| 1949 |
+
"grad_norm": 2.645836114883423,
|
| 1950 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 1951 |
+
"loss": 0.6626,
|
| 1952 |
+
"step": 61000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 78.40616966580977,
|
| 1956 |
+
"eval_accuracy": 0.8128617042506485,
|
| 1957 |
+
"eval_loss": 0.8741999268531799,
|
| 1958 |
+
"eval_runtime": 32.0443,
|
| 1959 |
+
"eval_samples_per_second": 194.2,
|
| 1960 |
+
"eval_steps_per_second": 6.085,
|
| 1961 |
+
"step": 61000
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 79.04884318766067,
|
| 1965 |
+
"grad_norm": 2.858215570449829,
|
| 1966 |
+
"learning_rate": 3.85e-05,
|
| 1967 |
+
"loss": 0.6617,
|
| 1968 |
+
"step": 61500
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 79.04884318766067,
|
| 1972 |
+
"eval_accuracy": 0.8136581954693509,
|
| 1973 |
+
"eval_loss": 0.8712067604064941,
|
| 1974 |
+
"eval_runtime": 32.061,
|
| 1975 |
+
"eval_samples_per_second": 194.099,
|
| 1976 |
+
"eval_steps_per_second": 6.082,
|
| 1977 |
+
"step": 61500
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 79.69151670951157,
|
| 1981 |
+
"grad_norm": 2.5988807678222656,
|
| 1982 |
+
"learning_rate": 3.8e-05,
|
| 1983 |
+
"loss": 0.655,
|
| 1984 |
+
"step": 62000
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 79.69151670951157,
|
| 1988 |
+
"eval_accuracy": 0.8128823572620334,
|
| 1989 |
+
"eval_loss": 0.8742080926895142,
|
| 1990 |
+
"eval_runtime": 32.0565,
|
| 1991 |
+
"eval_samples_per_second": 194.126,
|
| 1992 |
+
"eval_steps_per_second": 6.083,
|
| 1993 |
+
"step": 62000
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 80.33419023136247,
|
| 1997 |
+
"grad_norm": 2.6401941776275635,
|
| 1998 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 1999 |
+
"loss": 0.6545,
|
| 2000 |
+
"step": 62500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 80.33419023136247,
|
| 2004 |
+
"eval_accuracy": 0.812843042566724,
|
| 2005 |
+
"eval_loss": 0.8718605041503906,
|
| 2006 |
+
"eval_runtime": 31.9943,
|
| 2007 |
+
"eval_samples_per_second": 194.503,
|
| 2008 |
+
"eval_steps_per_second": 6.095,
|
| 2009 |
+
"step": 62500
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 80.97686375321337,
|
| 2013 |
+
"grad_norm": 2.8062636852264404,
|
| 2014 |
+
"learning_rate": 3.7e-05,
|
| 2015 |
+
"loss": 0.655,
|
| 2016 |
+
"step": 63000
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 80.97686375321337,
|
| 2020 |
+
"eval_accuracy": 0.811865863240631,
|
| 2021 |
+
"eval_loss": 0.8775948882102966,
|
| 2022 |
+
"eval_runtime": 32.0581,
|
| 2023 |
+
"eval_samples_per_second": 194.116,
|
| 2024 |
+
"eval_steps_per_second": 6.083,
|
| 2025 |
+
"step": 63000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 81.61953727506426,
|
| 2029 |
+
"grad_norm": 2.548220157623291,
|
| 2030 |
+
"learning_rate": 3.65e-05,
|
| 2031 |
+
"loss": 0.6468,
|
| 2032 |
+
"step": 63500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 81.61953727506426,
|
| 2036 |
+
"eval_accuracy": 0.8119920345795418,
|
| 2037 |
+
"eval_loss": 0.883579432964325,
|
| 2038 |
+
"eval_runtime": 32.0753,
|
| 2039 |
+
"eval_samples_per_second": 194.012,
|
| 2040 |
+
"eval_steps_per_second": 6.079,
|
| 2041 |
+
"step": 63500
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 82.26221079691517,
|
| 2045 |
+
"grad_norm": 2.5861518383026123,
|
| 2046 |
+
"learning_rate": 3.6e-05,
|
| 2047 |
+
"loss": 0.6487,
|
| 2048 |
+
"step": 64000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 82.26221079691517,
|
| 2052 |
+
"eval_accuracy": 0.8131239363417762,
|
| 2053 |
+
"eval_loss": 0.8730902075767517,
|
| 2054 |
+
"eval_runtime": 32.1045,
|
| 2055 |
+
"eval_samples_per_second": 193.835,
|
| 2056 |
+
"eval_steps_per_second": 6.074,
|
| 2057 |
+
"step": 64000
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 82.90488431876607,
|
| 2061 |
+
"grad_norm": 2.780217409133911,
|
| 2062 |
+
"learning_rate": 3.55e-05,
|
| 2063 |
+
"loss": 0.6495,
|
| 2064 |
+
"step": 64500
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 82.90488431876607,
|
| 2068 |
+
"eval_accuracy": 0.8124547391400174,
|
| 2069 |
+
"eval_loss": 0.8819155097007751,
|
| 2070 |
+
"eval_runtime": 32.0183,
|
| 2071 |
+
"eval_samples_per_second": 194.357,
|
| 2072 |
+
"eval_steps_per_second": 6.09,
|
| 2073 |
+
"step": 64500
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 83.54755784061696,
|
| 2077 |
+
"grad_norm": 2.5522782802581787,
|
| 2078 |
+
"learning_rate": 3.5e-05,
|
| 2079 |
+
"loss": 0.6425,
|
| 2080 |
+
"step": 65000
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 83.54755784061696,
|
| 2084 |
+
"eval_accuracy": 0.8124381288595807,
|
| 2085 |
+
"eval_loss": 0.885295569896698,
|
| 2086 |
+
"eval_runtime": 32.0293,
|
| 2087 |
+
"eval_samples_per_second": 194.291,
|
| 2088 |
+
"eval_steps_per_second": 6.088,
|
| 2089 |
+
"step": 65000
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 84.19023136246787,
|
| 2093 |
+
"grad_norm": 2.6785833835601807,
|
| 2094 |
+
"learning_rate": 3.45e-05,
|
| 2095 |
+
"loss": 0.6423,
|
| 2096 |
+
"step": 65500
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 84.19023136246787,
|
| 2100 |
+
"eval_accuracy": 0.8132641437128263,
|
| 2101 |
+
"eval_loss": 0.8791692852973938,
|
| 2102 |
+
"eval_runtime": 32.0427,
|
| 2103 |
+
"eval_samples_per_second": 194.209,
|
| 2104 |
+
"eval_steps_per_second": 6.086,
|
| 2105 |
+
"step": 65500
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 84.83290488431876,
|
| 2109 |
+
"grad_norm": 2.6489408016204834,
|
| 2110 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 2111 |
+
"loss": 0.6377,
|
| 2112 |
+
"step": 66000
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 84.83290488431876,
|
| 2116 |
+
"eval_accuracy": 0.8154822401853656,
|
| 2117 |
+
"eval_loss": 0.8657127618789673,
|
| 2118 |
+
"eval_runtime": 32.102,
|
| 2119 |
+
"eval_samples_per_second": 193.851,
|
| 2120 |
+
"eval_steps_per_second": 6.074,
|
| 2121 |
+
"step": 66000
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 85.47557840616966,
|
| 2125 |
+
"grad_norm": 2.556199312210083,
|
| 2126 |
+
"learning_rate": 3.35e-05,
|
| 2127 |
+
"loss": 0.6334,
|
| 2128 |
+
"step": 66500
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 85.47557840616966,
|
| 2132 |
+
"eval_accuracy": 0.8141076970581924,
|
| 2133 |
+
"eval_loss": 0.8747490644454956,
|
| 2134 |
+
"eval_runtime": 32.0636,
|
| 2135 |
+
"eval_samples_per_second": 194.083,
|
| 2136 |
+
"eval_steps_per_second": 6.082,
|
| 2137 |
+
"step": 66500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 86.11825192802057,
|
| 2141 |
+
"grad_norm": 2.477529764175415,
|
| 2142 |
+
"learning_rate": 3.3e-05,
|
| 2143 |
+
"loss": 0.6421,
|
| 2144 |
+
"step": 67000
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 86.11825192802057,
|
| 2148 |
+
"eval_accuracy": 0.8136067662347515,
|
| 2149 |
+
"eval_loss": 0.8816725015640259,
|
| 2150 |
+
"eval_runtime": 32.0966,
|
| 2151 |
+
"eval_samples_per_second": 193.883,
|
| 2152 |
+
"eval_steps_per_second": 6.075,
|
| 2153 |
+
"step": 67000
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 86.76092544987146,
|
| 2157 |
+
"grad_norm": 2.6774518489837646,
|
| 2158 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 2159 |
+
"loss": 0.6366,
|
| 2160 |
+
"step": 67500
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 86.76092544987146,
|
| 2164 |
+
"eval_accuracy": 0.814412672148727,
|
| 2165 |
+
"eval_loss": 0.8695808053016663,
|
| 2166 |
+
"eval_runtime": 32.0613,
|
| 2167 |
+
"eval_samples_per_second": 194.097,
|
| 2168 |
+
"eval_steps_per_second": 6.082,
|
| 2169 |
+
"step": 67500
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 87.40359897172236,
|
| 2173 |
+
"grad_norm": 2.6455330848693848,
|
| 2174 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 2175 |
+
"loss": 0.6305,
|
| 2176 |
+
"step": 68000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 87.40359897172236,
|
| 2180 |
+
"eval_accuracy": 0.8148675345064433,
|
| 2181 |
+
"eval_loss": 0.8716031312942505,
|
| 2182 |
+
"eval_runtime": 32.0969,
|
| 2183 |
+
"eval_samples_per_second": 193.882,
|
| 2184 |
+
"eval_steps_per_second": 6.075,
|
| 2185 |
+
"step": 68000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 88.04627249357327,
|
| 2189 |
+
"grad_norm": 2.7725086212158203,
|
| 2190 |
+
"learning_rate": 3.15e-05,
|
| 2191 |
+
"loss": 0.6325,
|
| 2192 |
+
"step": 68500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 88.04627249357327,
|
| 2196 |
+
"eval_accuracy": 0.813131776644422,
|
| 2197 |
+
"eval_loss": 0.8786645531654358,
|
| 2198 |
+
"eval_runtime": 32.024,
|
| 2199 |
+
"eval_samples_per_second": 194.323,
|
| 2200 |
+
"eval_steps_per_second": 6.089,
|
| 2201 |
+
"step": 68500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 88.68894601542416,
|
| 2205 |
+
"grad_norm": 2.6008856296539307,
|
| 2206 |
+
"learning_rate": 3.1e-05,
|
| 2207 |
+
"loss": 0.6282,
|
| 2208 |
+
"step": 69000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 88.68894601542416,
|
| 2212 |
+
"eval_accuracy": 0.8146181108967369,
|
| 2213 |
+
"eval_loss": 0.8678516149520874,
|
| 2214 |
+
"eval_runtime": 32.0037,
|
| 2215 |
+
"eval_samples_per_second": 194.447,
|
| 2216 |
+
"eval_steps_per_second": 6.093,
|
| 2217 |
+
"step": 69000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 89.33161953727506,
|
| 2221 |
+
"grad_norm": 2.7128331661224365,
|
| 2222 |
+
"learning_rate": 3.05e-05,
|
| 2223 |
+
"loss": 0.6261,
|
| 2224 |
+
"step": 69500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 89.33161953727506,
|
| 2228 |
+
"eval_accuracy": 0.8144260367762641,
|
| 2229 |
+
"eval_loss": 0.871059000492096,
|
| 2230 |
+
"eval_runtime": 32.0173,
|
| 2231 |
+
"eval_samples_per_second": 194.364,
|
| 2232 |
+
"eval_steps_per_second": 6.09,
|
| 2233 |
+
"step": 69500
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 89.97429305912597,
|
| 2237 |
+
"grad_norm": 2.324878692626953,
|
| 2238 |
+
"learning_rate": 3e-05,
|
| 2239 |
+
"loss": 0.6263,
|
| 2240 |
+
"step": 70000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 89.97429305912597,
|
| 2244 |
+
"eval_accuracy": 0.8156161510225485,
|
| 2245 |
+
"eval_loss": 0.8654680252075195,
|
| 2246 |
+
"eval_runtime": 32.0295,
|
| 2247 |
+
"eval_samples_per_second": 194.29,
|
| 2248 |
+
"eval_steps_per_second": 6.088,
|
| 2249 |
+
"step": 70000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 90.61696658097686,
|
| 2253 |
+
"grad_norm": 2.7399654388427734,
|
| 2254 |
+
"learning_rate": 2.95e-05,
|
| 2255 |
+
"loss": 0.6233,
|
| 2256 |
+
"step": 70500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 90.61696658097686,
|
| 2260 |
+
"eval_accuracy": 0.8162172670337523,
|
| 2261 |
+
"eval_loss": 0.8661888241767883,
|
| 2262 |
+
"eval_runtime": 32.0203,
|
| 2263 |
+
"eval_samples_per_second": 194.346,
|
| 2264 |
+
"eval_steps_per_second": 6.09,
|
| 2265 |
+
"step": 70500
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 91.25964010282776,
|
| 2269 |
+
"grad_norm": 2.8856687545776367,
|
| 2270 |
+
"learning_rate": 2.9e-05,
|
| 2271 |
+
"loss": 0.6176,
|
| 2272 |
+
"step": 71000
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 91.25964010282776,
|
| 2276 |
+
"eval_accuracy": 0.8161184520753214,
|
| 2277 |
+
"eval_loss": 0.8683423399925232,
|
| 2278 |
+
"eval_runtime": 32.0579,
|
| 2279 |
+
"eval_samples_per_second": 194.117,
|
| 2280 |
+
"eval_steps_per_second": 6.083,
|
| 2281 |
+
"step": 71000
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 91.90231362467867,
|
| 2285 |
+
"grad_norm": 2.4138548374176025,
|
| 2286 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 2287 |
+
"loss": 0.6239,
|
| 2288 |
+
"step": 71500
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 91.90231362467867,
|
| 2292 |
+
"eval_accuracy": 0.8153887598602306,
|
| 2293 |
+
"eval_loss": 0.8599680066108704,
|
| 2294 |
+
"eval_runtime": 32.0297,
|
| 2295 |
+
"eval_samples_per_second": 194.289,
|
| 2296 |
+
"eval_steps_per_second": 6.088,
|
| 2297 |
+
"step": 71500
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 92.54498714652956,
|
| 2301 |
+
"grad_norm": 2.6343512535095215,
|
| 2302 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 2303 |
+
"loss": 0.6189,
|
| 2304 |
+
"step": 72000
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 92.54498714652956,
|
| 2308 |
+
"eval_accuracy": 0.8154344802246543,
|
| 2309 |
+
"eval_loss": 0.8673732876777649,
|
| 2310 |
+
"eval_runtime": 32.0317,
|
| 2311 |
+
"eval_samples_per_second": 194.276,
|
| 2312 |
+
"eval_steps_per_second": 6.088,
|
| 2313 |
+
"step": 72000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 93.18766066838046,
|
| 2317 |
+
"grad_norm": 2.6886157989501953,
|
| 2318 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 2319 |
+
"loss": 0.6117,
|
| 2320 |
+
"step": 72500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 93.18766066838046,
|
| 2324 |
+
"eval_accuracy": 0.8162594874166894,
|
| 2325 |
+
"eval_loss": 0.8670706748962402,
|
| 2326 |
+
"eval_runtime": 32.0796,
|
| 2327 |
+
"eval_samples_per_second": 193.986,
|
| 2328 |
+
"eval_steps_per_second": 6.079,
|
| 2329 |
+
"step": 72500
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 93.83033419023137,
|
| 2333 |
+
"grad_norm": 2.6037755012512207,
|
| 2334 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 2335 |
+
"loss": 0.615,
|
| 2336 |
+
"step": 73000
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 93.83033419023137,
|
| 2340 |
+
"eval_accuracy": 0.8150874471307802,
|
| 2341 |
+
"eval_loss": 0.8663885593414307,
|
| 2342 |
+
"eval_runtime": 32.0301,
|
| 2343 |
+
"eval_samples_per_second": 194.286,
|
| 2344 |
+
"eval_steps_per_second": 6.088,
|
| 2345 |
+
"step": 73000
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 94.47300771208226,
|
| 2349 |
+
"grad_norm": 2.5215036869049072,
|
| 2350 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 2351 |
+
"loss": 0.6138,
|
| 2352 |
+
"step": 73500
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 94.47300771208226,
|
| 2356 |
+
"eval_accuracy": 0.8168731185953577,
|
| 2357 |
+
"eval_loss": 0.8630892634391785,
|
| 2358 |
+
"eval_runtime": 32.066,
|
| 2359 |
+
"eval_samples_per_second": 194.069,
|
| 2360 |
+
"eval_steps_per_second": 6.081,
|
| 2361 |
+
"step": 73500
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 95.11568123393316,
|
| 2365 |
+
"grad_norm": 2.797325611114502,
|
| 2366 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 2367 |
+
"loss": 0.6107,
|
| 2368 |
+
"step": 74000
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 95.11568123393316,
|
| 2372 |
+
"eval_accuracy": 0.8156241335474724,
|
| 2373 |
+
"eval_loss": 0.8718482851982117,
|
| 2374 |
+
"eval_runtime": 32.0559,
|
| 2375 |
+
"eval_samples_per_second": 194.13,
|
| 2376 |
+
"eval_steps_per_second": 6.083,
|
| 2377 |
+
"step": 74000
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 95.75835475578407,
|
| 2381 |
+
"grad_norm": 3.0883982181549072,
|
| 2382 |
+
"learning_rate": 2.5500000000000003e-05,
|
| 2383 |
+
"loss": 0.6099,
|
| 2384 |
+
"step": 74500
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 95.75835475578407,
|
| 2388 |
+
"eval_accuracy": 0.8160560285491784,
|
| 2389 |
+
"eval_loss": 0.8676818013191223,
|
| 2390 |
+
"eval_runtime": 32.0504,
|
| 2391 |
+
"eval_samples_per_second": 194.163,
|
| 2392 |
+
"eval_steps_per_second": 6.084,
|
| 2393 |
+
"step": 74500
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 96.40102827763496,
|
| 2397 |
+
"grad_norm": 2.922588348388672,
|
| 2398 |
+
"learning_rate": 2.5e-05,
|
| 2399 |
+
"loss": 0.6026,
|
| 2400 |
+
"step": 75000
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 96.40102827763496,
|
| 2404 |
+
"eval_accuracy": 0.8162867029734353,
|
| 2405 |
+
"eval_loss": 0.8721068501472473,
|
| 2406 |
+
"eval_runtime": 32.019,
|
| 2407 |
+
"eval_samples_per_second": 194.353,
|
| 2408 |
+
"eval_steps_per_second": 6.09,
|
| 2409 |
+
"step": 75000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 97.04370179948586,
|
| 2413 |
+
"grad_norm": 2.663724660873413,
|
| 2414 |
+
"learning_rate": 2.45e-05,
|
| 2415 |
+
"loss": 0.6084,
|
| 2416 |
+
"step": 75500
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 97.04370179948586,
|
| 2420 |
+
"eval_accuracy": 0.8182687960636296,
|
| 2421 |
+
"eval_loss": 0.858277440071106,
|
| 2422 |
+
"eval_runtime": 32.0344,
|
| 2423 |
+
"eval_samples_per_second": 194.26,
|
| 2424 |
+
"eval_steps_per_second": 6.087,
|
| 2425 |
+
"step": 75500
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 97.68637532133675,
|
| 2429 |
+
"grad_norm": 2.8198440074920654,
|
| 2430 |
+
"learning_rate": 2.4e-05,
|
| 2431 |
+
"loss": 0.6033,
|
| 2432 |
+
"step": 76000
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 97.68637532133675,
|
| 2436 |
+
"eval_accuracy": 0.8163839511643823,
|
| 2437 |
+
"eval_loss": 0.8688974380493164,
|
| 2438 |
+
"eval_runtime": 32.0629,
|
| 2439 |
+
"eval_samples_per_second": 194.087,
|
| 2440 |
+
"eval_steps_per_second": 6.082,
|
| 2441 |
+
"step": 76000
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 98.32904884318766,
|
| 2445 |
+
"grad_norm": 2.3979711532592773,
|
| 2446 |
+
"learning_rate": 2.35e-05,
|
| 2447 |
+
"loss": 0.603,
|
| 2448 |
+
"step": 76500
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 98.32904884318766,
|
| 2452 |
+
"eval_accuracy": 0.8169446192573946,
|
| 2453 |
+
"eval_loss": 0.8678939342498779,
|
| 2454 |
+
"eval_runtime": 32.0422,
|
| 2455 |
+
"eval_samples_per_second": 194.213,
|
| 2456 |
+
"eval_steps_per_second": 6.086,
|
| 2457 |
+
"step": 76500
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"epoch": 98.97172236503856,
|
| 2461 |
+
"grad_norm": 2.643324613571167,
|
| 2462 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 2463 |
+
"loss": 0.6009,
|
| 2464 |
+
"step": 77000
|
| 2465 |
+
},
|
| 2466 |
+
{
|
| 2467 |
+
"epoch": 98.97172236503856,
|
| 2468 |
+
"eval_accuracy": 0.8176424081803114,
|
| 2469 |
+
"eval_loss": 0.8609100580215454,
|
| 2470 |
+
"eval_runtime": 32.0681,
|
| 2471 |
+
"eval_samples_per_second": 194.056,
|
| 2472 |
+
"eval_steps_per_second": 6.081,
|
| 2473 |
+
"step": 77000
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 99.61439588688945,
|
| 2477 |
+
"grad_norm": 2.791355848312378,
|
| 2478 |
+
"learning_rate": 2.25e-05,
|
| 2479 |
+
"loss": 0.601,
|
| 2480 |
+
"step": 77500
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 99.61439588688945,
|
| 2484 |
+
"eval_accuracy": 0.816467387038539,
|
| 2485 |
+
"eval_loss": 0.867489755153656,
|
| 2486 |
+
"eval_runtime": 32.1192,
|
| 2487 |
+
"eval_samples_per_second": 193.747,
|
| 2488 |
+
"eval_steps_per_second": 6.071,
|
| 2489 |
+
"step": 77500
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 100.25706940874036,
|
| 2493 |
+
"grad_norm": 2.787290096282959,
|
| 2494 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 2495 |
+
"loss": 0.5966,
|
| 2496 |
+
"step": 78000
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 100.25706940874036,
|
| 2500 |
+
"eval_accuracy": 0.8171603328039044,
|
| 2501 |
+
"eval_loss": 0.8652510046958923,
|
| 2502 |
+
"eval_runtime": 32.0539,
|
| 2503 |
+
"eval_samples_per_second": 194.142,
|
| 2504 |
+
"eval_steps_per_second": 6.083,
|
| 2505 |
+
"step": 78000
|
| 2506 |
+
},
|
| 2507 |
+
{
|
| 2508 |
+
"epoch": 100.89974293059126,
|
| 2509 |
+
"grad_norm": 2.791646957397461,
|
| 2510 |
+
"learning_rate": 2.15e-05,
|
| 2511 |
+
"loss": 0.5935,
|
| 2512 |
+
"step": 78500
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 100.89974293059126,
|
| 2516 |
+
"eval_accuracy": 0.8178409455191369,
|
| 2517 |
+
"eval_loss": 0.8600234985351562,
|
| 2518 |
+
"eval_runtime": 32.0149,
|
| 2519 |
+
"eval_samples_per_second": 194.378,
|
| 2520 |
+
"eval_steps_per_second": 6.091,
|
| 2521 |
+
"step": 78500
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 101.54241645244215,
|
| 2525 |
+
"grad_norm": 2.4990010261535645,
|
| 2526 |
+
"learning_rate": 2.1e-05,
|
| 2527 |
+
"loss": 0.5912,
|
| 2528 |
+
"step": 79000
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 101.54241645244215,
|
| 2532 |
+
"eval_accuracy": 0.8169466975666281,
|
| 2533 |
+
"eval_loss": 0.8653005957603455,
|
| 2534 |
+
"eval_runtime": 31.9972,
|
| 2535 |
+
"eval_samples_per_second": 194.486,
|
| 2536 |
+
"eval_steps_per_second": 6.094,
|
| 2537 |
+
"step": 79000
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 102.18508997429306,
|
| 2541 |
+
"grad_norm": 2.3002467155456543,
|
| 2542 |
+
"learning_rate": 2.05e-05,
|
| 2543 |
+
"loss": 0.5873,
|
| 2544 |
+
"step": 79500
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 102.18508997429306,
|
| 2548 |
+
"eval_accuracy": 0.81857837810817,
|
| 2549 |
+
"eval_loss": 0.8576123118400574,
|
| 2550 |
+
"eval_runtime": 31.9904,
|
| 2551 |
+
"eval_samples_per_second": 194.527,
|
| 2552 |
+
"eval_steps_per_second": 6.096,
|
| 2553 |
+
"step": 79500
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 102.82776349614396,
|
| 2557 |
+
"grad_norm": 2.712336540222168,
|
| 2558 |
+
"learning_rate": 2e-05,
|
| 2559 |
+
"loss": 0.5913,
|
| 2560 |
+
"step": 80000
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 102.82776349614396,
|
| 2564 |
+
"eval_accuracy": 0.8185828450767534,
|
| 2565 |
+
"eval_loss": 0.855664074420929,
|
| 2566 |
+
"eval_runtime": 32.0488,
|
| 2567 |
+
"eval_samples_per_second": 194.173,
|
| 2568 |
+
"eval_steps_per_second": 6.084,
|
| 2569 |
+
"step": 80000
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 103.47043701799485,
|
| 2573 |
+
"grad_norm": 2.769562005996704,
|
| 2574 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 2575 |
+
"loss": 0.5908,
|
| 2576 |
+
"step": 80500
|
| 2577 |
+
},
|
| 2578 |
+
{
|
| 2579 |
+
"epoch": 103.47043701799485,
|
| 2580 |
+
"eval_accuracy": 0.8184711461830915,
|
| 2581 |
+
"eval_loss": 0.8605988621711731,
|
| 2582 |
+
"eval_runtime": 32.0286,
|
| 2583 |
+
"eval_samples_per_second": 194.295,
|
| 2584 |
+
"eval_steps_per_second": 6.088,
|
| 2585 |
+
"step": 80500
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 104.11311053984576,
|
| 2589 |
+
"grad_norm": 2.6929941177368164,
|
| 2590 |
+
"learning_rate": 1.9e-05,
|
| 2591 |
+
"loss": 0.5898,
|
| 2592 |
+
"step": 81000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 104.11311053984576,
|
| 2596 |
+
"eval_accuracy": 0.8183153166169729,
|
| 2597 |
+
"eval_loss": 0.8647379875183105,
|
| 2598 |
+
"eval_runtime": 32.0673,
|
| 2599 |
+
"eval_samples_per_second": 194.06,
|
| 2600 |
+
"eval_steps_per_second": 6.081,
|
| 2601 |
+
"step": 81000
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 104.75578406169666,
|
| 2605 |
+
"grad_norm": 2.8258962631225586,
|
| 2606 |
+
"learning_rate": 1.85e-05,
|
| 2607 |
+
"loss": 0.5841,
|
| 2608 |
+
"step": 81500
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 104.75578406169666,
|
| 2612 |
+
"eval_accuracy": 0.817774798787225,
|
| 2613 |
+
"eval_loss": 0.8651922941207886,
|
| 2614 |
+
"eval_runtime": 32.011,
|
| 2615 |
+
"eval_samples_per_second": 194.402,
|
| 2616 |
+
"eval_steps_per_second": 6.092,
|
| 2617 |
+
"step": 81500
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 105.39845758354755,
|
| 2621 |
+
"grad_norm": 2.46081805229187,
|
| 2622 |
+
"learning_rate": 1.8e-05,
|
| 2623 |
+
"loss": 0.582,
|
| 2624 |
+
"step": 82000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 105.39845758354755,
|
| 2628 |
+
"eval_accuracy": 0.817711216701606,
|
| 2629 |
+
"eval_loss": 0.8619120717048645,
|
| 2630 |
+
"eval_runtime": 32.0215,
|
| 2631 |
+
"eval_samples_per_second": 194.338,
|
| 2632 |
+
"eval_steps_per_second": 6.09,
|
| 2633 |
+
"step": 82000
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 106.04113110539846,
|
| 2637 |
+
"grad_norm": 2.960406541824341,
|
| 2638 |
+
"learning_rate": 1.75e-05,
|
| 2639 |
+
"loss": 0.5833,
|
| 2640 |
+
"step": 82500
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 106.04113110539846,
|
| 2644 |
+
"eval_accuracy": 0.8198030559915176,
|
| 2645 |
+
"eval_loss": 0.8566803932189941,
|
| 2646 |
+
"eval_runtime": 31.9884,
|
| 2647 |
+
"eval_samples_per_second": 194.539,
|
| 2648 |
+
"eval_steps_per_second": 6.096,
|
| 2649 |
+
"step": 82500
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 106.68380462724936,
|
| 2653 |
+
"grad_norm": 2.3195433616638184,
|
| 2654 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 2655 |
+
"loss": 0.5804,
|
| 2656 |
+
"step": 83000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 106.68380462724936,
|
| 2660 |
+
"eval_accuracy": 0.8193473285144606,
|
| 2661 |
+
"eval_loss": 0.8546783328056335,
|
| 2662 |
+
"eval_runtime": 32.0522,
|
| 2663 |
+
"eval_samples_per_second": 194.152,
|
| 2664 |
+
"eval_steps_per_second": 6.084,
|
| 2665 |
+
"step": 83000
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 107.32647814910025,
|
| 2669 |
+
"grad_norm": 2.7828962802886963,
|
| 2670 |
+
"learning_rate": 1.65e-05,
|
| 2671 |
+
"loss": 0.5839,
|
| 2672 |
+
"step": 83500
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 107.32647814910025,
|
| 2676 |
+
"eval_accuracy": 0.817869892847695,
|
| 2677 |
+
"eval_loss": 0.859340488910675,
|
| 2678 |
+
"eval_runtime": 32.2415,
|
| 2679 |
+
"eval_samples_per_second": 193.012,
|
| 2680 |
+
"eval_steps_per_second": 6.048,
|
| 2681 |
+
"step": 83500
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 107.96915167095116,
|
| 2685 |
+
"grad_norm": 2.8258416652679443,
|
| 2686 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 2687 |
+
"loss": 0.5774,
|
| 2688 |
+
"step": 84000
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 107.96915167095116,
|
| 2692 |
+
"eval_accuracy": 0.8181858662056681,
|
| 2693 |
+
"eval_loss": 0.8593936562538147,
|
| 2694 |
+
"eval_runtime": 32.0552,
|
| 2695 |
+
"eval_samples_per_second": 194.134,
|
| 2696 |
+
"eval_steps_per_second": 6.083,
|
| 2697 |
+
"step": 84000
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 108.61182519280206,
|
| 2701 |
+
"grad_norm": 2.5473225116729736,
|
| 2702 |
+
"learning_rate": 1.55e-05,
|
| 2703 |
+
"loss": 0.5791,
|
| 2704 |
+
"step": 84500
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 108.61182519280206,
|
| 2708 |
+
"eval_accuracy": 0.8201209689038143,
|
| 2709 |
+
"eval_loss": 0.8578335046768188,
|
| 2710 |
+
"eval_runtime": 32.0789,
|
| 2711 |
+
"eval_samples_per_second": 193.99,
|
| 2712 |
+
"eval_steps_per_second": 6.079,
|
| 2713 |
+
"step": 84500
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 109.25449871465295,
|
| 2717 |
+
"grad_norm": 2.6014022827148438,
|
| 2718 |
+
"learning_rate": 1.5e-05,
|
| 2719 |
+
"loss": 0.5724,
|
| 2720 |
+
"step": 85000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 109.25449871465295,
|
| 2724 |
+
"eval_accuracy": 0.8179463406457561,
|
| 2725 |
+
"eval_loss": 0.8615684509277344,
|
| 2726 |
+
"eval_runtime": 32.0619,
|
| 2727 |
+
"eval_samples_per_second": 194.093,
|
| 2728 |
+
"eval_steps_per_second": 6.082,
|
| 2729 |
+
"step": 85000
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 109.89717223650386,
|
| 2733 |
+
"grad_norm": 2.6044790744781494,
|
| 2734 |
+
"learning_rate": 1.45e-05,
|
| 2735 |
+
"loss": 0.5782,
|
| 2736 |
+
"step": 85500
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 109.89717223650386,
|
| 2740 |
+
"eval_accuracy": 0.818407341792689,
|
| 2741 |
+
"eval_loss": 0.8595022559165955,
|
| 2742 |
+
"eval_runtime": 32.0145,
|
| 2743 |
+
"eval_samples_per_second": 194.381,
|
| 2744 |
+
"eval_steps_per_second": 6.091,
|
| 2745 |
+
"step": 85500
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 110.53984575835476,
|
| 2749 |
+
"grad_norm": 2.570466995239258,
|
| 2750 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 2751 |
+
"loss": 0.5735,
|
| 2752 |
+
"step": 86000
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 110.53984575835476,
|
| 2756 |
+
"eval_accuracy": 0.8203102481461101,
|
| 2757 |
+
"eval_loss": 0.8511986136436462,
|
| 2758 |
+
"eval_runtime": 32.0544,
|
| 2759 |
+
"eval_samples_per_second": 194.139,
|
| 2760 |
+
"eval_steps_per_second": 6.083,
|
| 2761 |
+
"step": 86000
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 111.18251928020565,
|
| 2765 |
+
"grad_norm": 2.4701082706451416,
|
| 2766 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 2767 |
+
"loss": 0.5664,
|
| 2768 |
+
"step": 86500
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 111.18251928020565,
|
| 2772 |
+
"eval_accuracy": 0.8206346877390821,
|
| 2773 |
+
"eval_loss": 0.8540939688682556,
|
| 2774 |
+
"eval_runtime": 32.046,
|
| 2775 |
+
"eval_samples_per_second": 194.19,
|
| 2776 |
+
"eval_steps_per_second": 6.085,
|
| 2777 |
+
"step": 86500
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 111.82519280205655,
|
| 2781 |
+
"grad_norm": 2.9233691692352295,
|
| 2782 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 2783 |
+
"loss": 0.5719,
|
| 2784 |
+
"step": 87000
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 111.82519280205655,
|
| 2788 |
+
"eval_accuracy": 0.8200771187996747,
|
| 2789 |
+
"eval_loss": 0.8584414124488831,
|
| 2790 |
+
"eval_runtime": 32.0359,
|
| 2791 |
+
"eval_samples_per_second": 194.251,
|
| 2792 |
+
"eval_steps_per_second": 6.087,
|
| 2793 |
+
"step": 87000
|
| 2794 |
+
},
|
| 2795 |
+
{
|
| 2796 |
+
"epoch": 112.46786632390746,
|
| 2797 |
+
"grad_norm": 2.3986382484436035,
|
| 2798 |
+
"learning_rate": 1.25e-05,
|
| 2799 |
+
"loss": 0.5682,
|
| 2800 |
+
"step": 87500
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 112.46786632390746,
|
| 2804 |
+
"eval_accuracy": 0.8193907912255869,
|
| 2805 |
+
"eval_loss": 0.8565849661827087,
|
| 2806 |
+
"eval_runtime": 32.0525,
|
| 2807 |
+
"eval_samples_per_second": 194.15,
|
| 2808 |
+
"eval_steps_per_second": 6.084,
|
| 2809 |
+
"step": 87500
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 113.11053984575835,
|
| 2813 |
+
"grad_norm": 2.5226054191589355,
|
| 2814 |
+
"learning_rate": 1.2e-05,
|
| 2815 |
+
"loss": 0.5719,
|
| 2816 |
+
"step": 88000
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 113.11053984575835,
|
| 2820 |
+
"eval_accuracy": 0.8199272051177008,
|
| 2821 |
+
"eval_loss": 0.8500083684921265,
|
| 2822 |
+
"eval_runtime": 32.0889,
|
| 2823 |
+
"eval_samples_per_second": 193.93,
|
| 2824 |
+
"eval_steps_per_second": 6.077,
|
| 2825 |
+
"step": 88000
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 113.75321336760925,
|
| 2829 |
+
"grad_norm": 2.609858274459839,
|
| 2830 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 2831 |
+
"loss": 0.5674,
|
| 2832 |
+
"step": 88500
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 113.75321336760925,
|
| 2836 |
+
"eval_accuracy": 0.8205847363486674,
|
| 2837 |
+
"eval_loss": 0.8524363040924072,
|
| 2838 |
+
"eval_runtime": 32.0335,
|
| 2839 |
+
"eval_samples_per_second": 194.265,
|
| 2840 |
+
"eval_steps_per_second": 6.087,
|
| 2841 |
+
"step": 88500
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 114.39588688946016,
|
| 2845 |
+
"grad_norm": 2.2834887504577637,
|
| 2846 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 2847 |
+
"loss": 0.5643,
|
| 2848 |
+
"step": 89000
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 114.39588688946016,
|
| 2852 |
+
"eval_accuracy": 0.821479995279122,
|
| 2853 |
+
"eval_loss": 0.8454416990280151,
|
| 2854 |
+
"eval_runtime": 32.0112,
|
| 2855 |
+
"eval_samples_per_second": 194.401,
|
| 2856 |
+
"eval_steps_per_second": 6.092,
|
| 2857 |
+
"step": 89000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 115.03856041131105,
|
| 2861 |
+
"grad_norm": 2.776373863220215,
|
| 2862 |
+
"learning_rate": 1.05e-05,
|
| 2863 |
+
"loss": 0.5681,
|
| 2864 |
+
"step": 89500
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 115.03856041131105,
|
| 2868 |
+
"eval_accuracy": 0.8196546091505466,
|
| 2869 |
+
"eval_loss": 0.8588145971298218,
|
| 2870 |
+
"eval_runtime": 31.2312,
|
| 2871 |
+
"eval_samples_per_second": 199.256,
|
| 2872 |
+
"eval_steps_per_second": 6.244,
|
| 2873 |
+
"step": 89500
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 115.68123393316195,
|
| 2877 |
+
"grad_norm": 2.7792084217071533,
|
| 2878 |
+
"learning_rate": 1e-05,
|
| 2879 |
+
"loss": 0.563,
|
| 2880 |
+
"step": 90000
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 115.68123393316195,
|
| 2884 |
+
"eval_accuracy": 0.8215367161011942,
|
| 2885 |
+
"eval_loss": 0.8472273945808411,
|
| 2886 |
+
"eval_runtime": 32.0865,
|
| 2887 |
+
"eval_samples_per_second": 193.944,
|
| 2888 |
+
"eval_steps_per_second": 6.077,
|
| 2889 |
+
"step": 90000
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 116.32390745501286,
|
| 2893 |
+
"grad_norm": 2.5021698474884033,
|
| 2894 |
+
"learning_rate": 9.5e-06,
|
| 2895 |
+
"loss": 0.5563,
|
| 2896 |
+
"step": 90500
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 116.32390745501286,
|
| 2900 |
+
"eval_accuracy": 0.8198419059293134,
|
| 2901 |
+
"eval_loss": 0.8576837778091431,
|
| 2902 |
+
"eval_runtime": 32.0421,
|
| 2903 |
+
"eval_samples_per_second": 194.214,
|
| 2904 |
+
"eval_steps_per_second": 6.086,
|
| 2905 |
+
"step": 90500
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 116.96658097686375,
|
| 2909 |
+
"grad_norm": 2.7690136432647705,
|
| 2910 |
+
"learning_rate": 9e-06,
|
| 2911 |
+
"loss": 0.5667,
|
| 2912 |
+
"step": 91000
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 116.96658097686375,
|
| 2916 |
+
"eval_accuracy": 0.8201571685631533,
|
| 2917 |
+
"eval_loss": 0.8592654466629028,
|
| 2918 |
+
"eval_runtime": 32.0402,
|
| 2919 |
+
"eval_samples_per_second": 194.225,
|
| 2920 |
+
"eval_steps_per_second": 6.086,
|
| 2921 |
+
"step": 91000
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 117.60925449871465,
|
| 2925 |
+
"grad_norm": 2.7082719802856445,
|
| 2926 |
+
"learning_rate": 8.500000000000002e-06,
|
| 2927 |
+
"loss": 0.5598,
|
| 2928 |
+
"step": 91500
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 117.60925449871465,
|
| 2932 |
+
"eval_accuracy": 0.8194785991809854,
|
| 2933 |
+
"eval_loss": 0.8624646067619324,
|
| 2934 |
+
"eval_runtime": 31.9751,
|
| 2935 |
+
"eval_samples_per_second": 194.62,
|
| 2936 |
+
"eval_steps_per_second": 6.098,
|
| 2937 |
+
"step": 91500
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 118.25192802056556,
|
| 2941 |
+
"grad_norm": 2.6081254482269287,
|
| 2942 |
+
"learning_rate": 8.000000000000001e-06,
|
| 2943 |
+
"loss": 0.5588,
|
| 2944 |
+
"step": 92000
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 118.25192802056556,
|
| 2948 |
+
"eval_accuracy": 0.8200990472992823,
|
| 2949 |
+
"eval_loss": 0.8558657169342041,
|
| 2950 |
+
"eval_runtime": 32.0709,
|
| 2951 |
+
"eval_samples_per_second": 194.039,
|
| 2952 |
+
"eval_steps_per_second": 6.08,
|
| 2953 |
+
"step": 92000
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 118.89460154241645,
|
| 2957 |
+
"grad_norm": 2.39003849029541,
|
| 2958 |
+
"learning_rate": 7.5e-06,
|
| 2959 |
+
"loss": 0.5573,
|
| 2960 |
+
"step": 92500
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 118.89460154241645,
|
| 2964 |
+
"eval_accuracy": 0.821544695178437,
|
| 2965 |
+
"eval_loss": 0.8529016375541687,
|
| 2966 |
+
"eval_runtime": 32.0688,
|
| 2967 |
+
"eval_samples_per_second": 194.052,
|
| 2968 |
+
"eval_steps_per_second": 6.081,
|
| 2969 |
+
"step": 92500
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 119.53727506426735,
|
| 2973 |
+
"grad_norm": 2.671393871307373,
|
| 2974 |
+
"learning_rate": 7.000000000000001e-06,
|
| 2975 |
+
"loss": 0.5583,
|
| 2976 |
+
"step": 93000
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 119.53727506426735,
|
| 2980 |
+
"eval_accuracy": 0.8202954907050746,
|
| 2981 |
+
"eval_loss": 0.8513576984405518,
|
| 2982 |
+
"eval_runtime": 32.0535,
|
| 2983 |
+
"eval_samples_per_second": 194.144,
|
| 2984 |
+
"eval_steps_per_second": 6.084,
|
| 2985 |
+
"step": 93000
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 120.17994858611826,
|
| 2989 |
+
"grad_norm": 2.8136119842529297,
|
| 2990 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 2991 |
+
"loss": 0.5599,
|
| 2992 |
+
"step": 93500
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 120.17994858611826,
|
| 2996 |
+
"eval_accuracy": 0.8210522728285633,
|
| 2997 |
+
"eval_loss": 0.8533715605735779,
|
| 2998 |
+
"eval_runtime": 32.0108,
|
| 2999 |
+
"eval_samples_per_second": 194.403,
|
| 3000 |
+
"eval_steps_per_second": 6.092,
|
| 3001 |
+
"step": 93500
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 120.82262210796915,
|
| 3005 |
+
"grad_norm": 2.5622072219848633,
|
| 3006 |
+
"learning_rate": 6e-06,
|
| 3007 |
+
"loss": 0.5538,
|
| 3008 |
+
"step": 94000
|
| 3009 |
+
},
|
| 3010 |
+
{
|
| 3011 |
+
"epoch": 120.82262210796915,
|
| 3012 |
+
"eval_accuracy": 0.8210836727544613,
|
| 3013 |
+
"eval_loss": 0.8567976951599121,
|
| 3014 |
+
"eval_runtime": 32.0891,
|
| 3015 |
+
"eval_samples_per_second": 193.929,
|
| 3016 |
+
"eval_steps_per_second": 6.077,
|
| 3017 |
+
"step": 94000
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 121.46529562982005,
|
| 3021 |
+
"grad_norm": 2.870851516723633,
|
| 3022 |
+
"learning_rate": 5.500000000000001e-06,
|
| 3023 |
+
"loss": 0.5544,
|
| 3024 |
+
"step": 94500
|
| 3025 |
+
},
|
| 3026 |
+
{
|
| 3027 |
+
"epoch": 121.46529562982005,
|
| 3028 |
+
"eval_accuracy": 0.8227272131920523,
|
| 3029 |
+
"eval_loss": 0.8419870138168335,
|
| 3030 |
+
"eval_runtime": 32.0445,
|
| 3031 |
+
"eval_samples_per_second": 194.199,
|
| 3032 |
+
"eval_steps_per_second": 6.085,
|
| 3033 |
+
"step": 94500
|
| 3034 |
+
}
|
| 3035 |
+
],
|
| 3036 |
+
"logging_steps": 500,
|
| 3037 |
+
"max_steps": 100000,
|
| 3038 |
+
"num_input_tokens_seen": 0,
|
| 3039 |
+
"num_train_epochs": 129,
|
| 3040 |
+
"save_steps": 500,
|
| 3041 |
+
"stateful_callbacks": {
|
| 3042 |
+
"TrainerControl": {
|
| 3043 |
+
"args": {
|
| 3044 |
+
"should_epoch_stop": false,
|
| 3045 |
+
"should_evaluate": false,
|
| 3046 |
+
"should_log": false,
|
| 3047 |
+
"should_save": true,
|
| 3048 |
+
"should_training_stop": false
|
| 3049 |
+
},
|
| 3050 |
+
"attributes": {}
|
| 3051 |
+
}
|
| 3052 |
+
},
|
| 3053 |
+
"total_flos": 7.979699566574305e+17,
|
| 3054 |
+
"train_batch_size": 16,
|
| 3055 |
+
"trial_name": null,
|
| 3056 |
+
"trial_params": null
|
| 3057 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91c79fb44304cdfed921db0457e092f59972f8f1b81277a47f64dec6d5ff116f
|
| 3 |
+
size 5176
|