End of training
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- README.md +70 -0
- config.json +26 -0
- model.safetensors +3 -0
- run-0/checkpoint-2616/config.json +26 -0
- run-0/checkpoint-2616/model.safetensors +3 -0
- run-0/checkpoint-2616/optimizer.pt +3 -0
- run-0/checkpoint-2616/rng_state.pth +3 -0
- run-0/checkpoint-2616/scheduler.pt +3 -0
- run-0/checkpoint-2616/special_tokens_map.json +7 -0
- run-0/checkpoint-2616/tokenizer.json +0 -0
- run-0/checkpoint-2616/tokenizer_config.json +55 -0
- run-0/checkpoint-2616/trainer_state.json +1913 -0
- run-0/checkpoint-2616/training_args.bin +3 -0
- run-0/checkpoint-2616/vocab.txt +0 -0
- run-1/checkpoint-654/config.json +26 -0
- run-1/checkpoint-654/model.safetensors +3 -0
- run-1/checkpoint-654/optimizer.pt +3 -0
- run-1/checkpoint-654/rng_state.pth +3 -0
- run-1/checkpoint-654/scheduler.pt +3 -0
- run-1/checkpoint-654/special_tokens_map.json +7 -0
- run-1/checkpoint-654/tokenizer.json +0 -0
- run-1/checkpoint-654/tokenizer_config.json +55 -0
- run-1/checkpoint-654/trainer_state.json +505 -0
- run-1/checkpoint-654/training_args.bin +3 -0
- run-1/checkpoint-654/vocab.txt +0 -0
- run-2/checkpoint-492/config.json +26 -0
- run-2/checkpoint-492/model.safetensors +3 -0
- run-2/checkpoint-492/optimizer.pt +3 -0
- run-2/checkpoint-492/rng_state.pth +3 -0
- run-2/checkpoint-492/scheduler.pt +3 -0
- run-2/checkpoint-492/special_tokens_map.json +7 -0
- run-2/checkpoint-492/tokenizer.json +0 -0
- run-2/checkpoint-492/tokenizer_config.json +55 -0
- run-2/checkpoint-492/trainer_state.json +417 -0
- run-2/checkpoint-492/training_args.bin +3 -0
- run-2/checkpoint-492/vocab.txt +0 -0
- run-3/checkpoint-328/config.json +26 -0
- run-3/checkpoint-328/model.safetensors +3 -0
- run-3/checkpoint-328/optimizer.pt +3 -0
- run-3/checkpoint-328/rng_state.pth +3 -0
- run-3/checkpoint-328/scheduler.pt +3 -0
- run-3/checkpoint-328/special_tokens_map.json +7 -0
- run-3/checkpoint-328/tokenizer.json +0 -0
- run-3/checkpoint-328/tokenizer_config.json +55 -0
- run-3/checkpoint-328/trainer_state.json +310 -0
- run-3/checkpoint-328/training_args.bin +3 -0
- run-3/checkpoint-328/vocab.txt +0 -0
- run-4/checkpoint-1308/config.json +26 -0
- run-4/checkpoint-1308/model.safetensors +3 -0
- run-4/checkpoint-1308/optimizer.pt +3 -0
README.md
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: distilbert-base-multilingual-cased
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
metrics:
|
8 |
+
- accuracy
|
9 |
+
- recall
|
10 |
+
- precision
|
11 |
+
- f1
|
12 |
+
model-index:
|
13 |
+
- name: distilbert-base-multilingual-cased-aoe-hyper
|
14 |
+
results: []
|
15 |
+
---
|
16 |
+
|
17 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
18 |
+
should probably proofread and complete it, then remove this comment. -->
|
19 |
+
|
20 |
+
# distilbert-base-multilingual-cased-aoe-hyper
|
21 |
+
|
22 |
+
This model is a fine-tuned version of [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased) on the None dataset.
|
23 |
+
It achieves the following results on the evaluation set:
|
24 |
+
- Loss: 0.4021
|
25 |
+
- Accuracy: 0.8799
|
26 |
+
- Recall: 0.8013
|
27 |
+
- Precision: 0.7246
|
28 |
+
- F1: 0.7610
|
29 |
+
|
30 |
+
## Model description
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Intended uses & limitations
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training and evaluation data
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Training procedure
|
43 |
+
|
44 |
+
### Training hyperparameters
|
45 |
+
|
46 |
+
The following hyperparameters were used during training:
|
47 |
+
- learning_rate: 6.292247538797816e-06
|
48 |
+
- train_batch_size: 8
|
49 |
+
- eval_batch_size: 16
|
50 |
+
- seed: 16
|
51 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
52 |
+
- lr_scheduler_type: linear
|
53 |
+
- num_epochs: 4
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | Precision | F1 |
|
58 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
|
59 |
+
| 0.2696 | 1.0 | 654 | 0.3423 | 0.8699 | 0.6090 | 0.7983 | 0.6909 |
|
60 |
+
| 0.4222 | 2.0 | 1308 | 0.3537 | 0.8814 | 0.7596 | 0.7476 | 0.7536 |
|
61 |
+
| 0.3135 | 3.0 | 1962 | 0.3992 | 0.8699 | 0.8205 | 0.6919 | 0.7507 |
|
62 |
+
| 0.2558 | 4.0 | 2616 | 0.4021 | 0.8799 | 0.8013 | 0.7246 | 0.7610 |
|
63 |
+
|
64 |
+
|
65 |
+
### Framework versions
|
66 |
+
|
67 |
+
- Transformers 4.46.2
|
68 |
+
- Pytorch 2.5.1+cu121
|
69 |
+
- Datasets 3.1.0
|
70 |
+
- Tokenizers 0.20.3
|
config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3f013b5e59583780ee83d863dc493644833237543dbc19d7d4b2efccf3c015b
|
3 |
+
size 541317368
|
run-0/checkpoint-2616/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
run-0/checkpoint-2616/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ebc20b909f12366cfaed536bd0051462d09f7d7ba971a1115a72a4830846b2c3
|
3 |
+
size 541317368
|
run-0/checkpoint-2616/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a958f1fb7a130e7304d810babbd1dfea988ff5ad807fba1ed7722fff09fcaaa2
|
3 |
+
size 1082696890
|
run-0/checkpoint-2616/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5720cc9c6b35d5b26684c1cc949c8b6e8ce7e9c35117a034212aa1075dd7544
|
3 |
+
size 14244
|
run-0/checkpoint-2616/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:72769ff360c7834055f7cea1cdd36cd37c87f9692499d5e60332b743c1e9765f
|
3 |
+
size 1064
|
run-0/checkpoint-2616/special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
run-0/checkpoint-2616/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-0/checkpoint-2616/tokenizer_config.json
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"100": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"101": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"102": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"103": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": false,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"mask_token": "[MASK]",
|
48 |
+
"model_max_length": 512,
|
49 |
+
"pad_token": "[PAD]",
|
50 |
+
"sep_token": "[SEP]",
|
51 |
+
"strip_accents": null,
|
52 |
+
"tokenize_chinese_chars": true,
|
53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
54 |
+
"unk_token": "[UNK]"
|
55 |
+
}
|
run-0/checkpoint-2616/trainer_state.json
ADDED
@@ -0,0 +1,1913 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8859984697781178,
|
3 |
+
"best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-0/checkpoint-2616",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2616,
|
7 |
+
"is_hyper_param_search": true,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01529051987767584,
|
13 |
+
"grad_norm": 1.6358972787857056,
|
14 |
+
"learning_rate": 9.349717786759033e-06,
|
15 |
+
"loss": 0.6796,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.03058103975535168,
|
20 |
+
"grad_norm": 1.7600407600402832,
|
21 |
+
"learning_rate": 9.321037670848728e-06,
|
22 |
+
"loss": 0.5945,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.045871559633027525,
|
27 |
+
"grad_norm": 2.3243937492370605,
|
28 |
+
"learning_rate": 9.292357554938425e-06,
|
29 |
+
"loss": 0.5386,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.06116207951070336,
|
34 |
+
"grad_norm": 2.1618645191192627,
|
35 |
+
"learning_rate": 9.263677439028121e-06,
|
36 |
+
"loss": 0.4348,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.0764525993883792,
|
41 |
+
"grad_norm": 4.011763572692871,
|
42 |
+
"learning_rate": 9.234997323117817e-06,
|
43 |
+
"loss": 0.4917,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.09174311926605505,
|
48 |
+
"grad_norm": 3.0631511211395264,
|
49 |
+
"learning_rate": 9.206317207207513e-06,
|
50 |
+
"loss": 0.4419,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.10703363914373089,
|
55 |
+
"grad_norm": 5.19169807434082,
|
56 |
+
"learning_rate": 9.177637091297209e-06,
|
57 |
+
"loss": 0.4885,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.12232415902140673,
|
62 |
+
"grad_norm": 3.2671306133270264,
|
63 |
+
"learning_rate": 9.148956975386905e-06,
|
64 |
+
"loss": 0.3778,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.13761467889908258,
|
69 |
+
"grad_norm": 2.0801641941070557,
|
70 |
+
"learning_rate": 9.120276859476602e-06,
|
71 |
+
"loss": 0.4524,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1529051987767584,
|
76 |
+
"grad_norm": 3.2077746391296387,
|
77 |
+
"learning_rate": 9.091596743566297e-06,
|
78 |
+
"loss": 0.3107,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.16819571865443425,
|
83 |
+
"grad_norm": 3.8301804065704346,
|
84 |
+
"learning_rate": 9.062916627655994e-06,
|
85 |
+
"loss": 0.4904,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.1834862385321101,
|
90 |
+
"grad_norm": 2.995657205581665,
|
91 |
+
"learning_rate": 9.03423651174569e-06,
|
92 |
+
"loss": 0.3564,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.19877675840978593,
|
97 |
+
"grad_norm": 5.345709800720215,
|
98 |
+
"learning_rate": 9.005556395835386e-06,
|
99 |
+
"loss": 0.498,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.21406727828746178,
|
104 |
+
"grad_norm": 1.6241793632507324,
|
105 |
+
"learning_rate": 8.976876279925083e-06,
|
106 |
+
"loss": 0.4324,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.22935779816513763,
|
111 |
+
"grad_norm": 6.096868991851807,
|
112 |
+
"learning_rate": 8.94819616401478e-06,
|
113 |
+
"loss": 0.4474,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.24464831804281345,
|
118 |
+
"grad_norm": 4.777408123016357,
|
119 |
+
"learning_rate": 8.919516048104475e-06,
|
120 |
+
"loss": 0.3605,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.2599388379204893,
|
125 |
+
"grad_norm": 4.883125305175781,
|
126 |
+
"learning_rate": 8.890835932194172e-06,
|
127 |
+
"loss": 0.3694,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.27522935779816515,
|
132 |
+
"grad_norm": 2.1104636192321777,
|
133 |
+
"learning_rate": 8.862155816283867e-06,
|
134 |
+
"loss": 0.3391,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.290519877675841,
|
139 |
+
"grad_norm": 3.5193943977355957,
|
140 |
+
"learning_rate": 8.833475700373564e-06,
|
141 |
+
"loss": 0.4028,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.3058103975535168,
|
146 |
+
"grad_norm": 12.695348739624023,
|
147 |
+
"learning_rate": 8.80479558446326e-06,
|
148 |
+
"loss": 0.5372,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.3211009174311927,
|
153 |
+
"grad_norm": 8.019516944885254,
|
154 |
+
"learning_rate": 8.776115468552956e-06,
|
155 |
+
"loss": 0.3839,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.3363914373088685,
|
160 |
+
"grad_norm": 2.723048448562622,
|
161 |
+
"learning_rate": 8.747435352642653e-06,
|
162 |
+
"loss": 0.3155,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.3516819571865443,
|
167 |
+
"grad_norm": 5.75403356552124,
|
168 |
+
"learning_rate": 8.718755236732348e-06,
|
169 |
+
"loss": 0.4086,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.3669724770642202,
|
174 |
+
"grad_norm": 3.325057029724121,
|
175 |
+
"learning_rate": 8.690075120822045e-06,
|
176 |
+
"loss": 0.293,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.382262996941896,
|
181 |
+
"grad_norm": 18.065032958984375,
|
182 |
+
"learning_rate": 8.66139500491174e-06,
|
183 |
+
"loss": 0.3126,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.39755351681957185,
|
188 |
+
"grad_norm": 5.72575569152832,
|
189 |
+
"learning_rate": 8.632714889001437e-06,
|
190 |
+
"loss": 0.3087,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.41284403669724773,
|
195 |
+
"grad_norm": 5.2425055503845215,
|
196 |
+
"learning_rate": 8.604034773091134e-06,
|
197 |
+
"loss": 0.474,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.42813455657492355,
|
202 |
+
"grad_norm": 8.025136947631836,
|
203 |
+
"learning_rate": 8.57535465718083e-06,
|
204 |
+
"loss": 0.2461,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.4434250764525994,
|
209 |
+
"grad_norm": 11.76362133026123,
|
210 |
+
"learning_rate": 8.546674541270526e-06,
|
211 |
+
"loss": 0.3092,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.45871559633027525,
|
216 |
+
"grad_norm": 14.632379531860352,
|
217 |
+
"learning_rate": 8.517994425360223e-06,
|
218 |
+
"loss": 0.4511,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.4740061162079511,
|
223 |
+
"grad_norm": 3.7113358974456787,
|
224 |
+
"learning_rate": 8.489314309449918e-06,
|
225 |
+
"loss": 0.3188,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.4892966360856269,
|
230 |
+
"grad_norm": 10.738109588623047,
|
231 |
+
"learning_rate": 8.460634193539615e-06,
|
232 |
+
"loss": 0.4563,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.5045871559633027,
|
237 |
+
"grad_norm": 13.628973960876465,
|
238 |
+
"learning_rate": 8.431954077629312e-06,
|
239 |
+
"loss": 0.3233,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.5198776758409785,
|
244 |
+
"grad_norm": 17.098241806030273,
|
245 |
+
"learning_rate": 8.403273961719007e-06,
|
246 |
+
"loss": 0.4134,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.5351681957186545,
|
251 |
+
"grad_norm": 4.145214080810547,
|
252 |
+
"learning_rate": 8.374593845808704e-06,
|
253 |
+
"loss": 0.3638,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.5504587155963303,
|
258 |
+
"grad_norm": 21.35369300842285,
|
259 |
+
"learning_rate": 8.345913729898399e-06,
|
260 |
+
"loss": 0.3341,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.5657492354740061,
|
265 |
+
"grad_norm": 6.4279937744140625,
|
266 |
+
"learning_rate": 8.317233613988096e-06,
|
267 |
+
"loss": 0.3332,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.581039755351682,
|
272 |
+
"grad_norm": 6.384445667266846,
|
273 |
+
"learning_rate": 8.288553498077791e-06,
|
274 |
+
"loss": 0.3299,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.5963302752293578,
|
279 |
+
"grad_norm": 9.927398681640625,
|
280 |
+
"learning_rate": 8.25987338216749e-06,
|
281 |
+
"loss": 0.3687,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.6116207951070336,
|
286 |
+
"grad_norm": 2.317152261734009,
|
287 |
+
"learning_rate": 8.231193266257185e-06,
|
288 |
+
"loss": 0.2947,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.6269113149847095,
|
293 |
+
"grad_norm": 6.1938982009887695,
|
294 |
+
"learning_rate": 8.202513150346881e-06,
|
295 |
+
"loss": 0.2495,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.6422018348623854,
|
300 |
+
"grad_norm": 9.2775297164917,
|
301 |
+
"learning_rate": 8.173833034436577e-06,
|
302 |
+
"loss": 0.3011,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.6574923547400612,
|
307 |
+
"grad_norm": 12.099929809570312,
|
308 |
+
"learning_rate": 8.145152918526273e-06,
|
309 |
+
"loss": 0.2866,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.672782874617737,
|
314 |
+
"grad_norm": 15.571517944335938,
|
315 |
+
"learning_rate": 8.116472802615969e-06,
|
316 |
+
"loss": 0.2352,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.6880733944954128,
|
321 |
+
"grad_norm": 2.7042486667633057,
|
322 |
+
"learning_rate": 8.087792686705666e-06,
|
323 |
+
"loss": 0.4442,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.7033639143730887,
|
328 |
+
"grad_norm": 24.800390243530273,
|
329 |
+
"learning_rate": 8.059112570795362e-06,
|
330 |
+
"loss": 0.2647,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.7186544342507645,
|
335 |
+
"grad_norm": 1.1386603116989136,
|
336 |
+
"learning_rate": 8.030432454885058e-06,
|
337 |
+
"loss": 0.339,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.7339449541284404,
|
342 |
+
"grad_norm": 2.4210026264190674,
|
343 |
+
"learning_rate": 8.001752338974754e-06,
|
344 |
+
"loss": 0.2963,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.7492354740061162,
|
349 |
+
"grad_norm": 10.266980171203613,
|
350 |
+
"learning_rate": 7.97307222306445e-06,
|
351 |
+
"loss": 0.2177,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.764525993883792,
|
356 |
+
"grad_norm": 8.048089027404785,
|
357 |
+
"learning_rate": 7.944392107154146e-06,
|
358 |
+
"loss": 0.3248,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.7798165137614679,
|
363 |
+
"grad_norm": 25.07794189453125,
|
364 |
+
"learning_rate": 7.915711991243843e-06,
|
365 |
+
"loss": 0.3321,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.7951070336391437,
|
370 |
+
"grad_norm": 25.960834503173828,
|
371 |
+
"learning_rate": 7.88703187533354e-06,
|
372 |
+
"loss": 0.4167,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.8103975535168195,
|
377 |
+
"grad_norm": 7.0256028175354,
|
378 |
+
"learning_rate": 7.858351759423235e-06,
|
379 |
+
"loss": 0.2602,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.8256880733944955,
|
384 |
+
"grad_norm": 3.3532984256744385,
|
385 |
+
"learning_rate": 7.829671643512932e-06,
|
386 |
+
"loss": 0.2569,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.8409785932721713,
|
391 |
+
"grad_norm": 5.293972969055176,
|
392 |
+
"learning_rate": 7.800991527602627e-06,
|
393 |
+
"loss": 0.321,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.8562691131498471,
|
398 |
+
"grad_norm": 5.642370223999023,
|
399 |
+
"learning_rate": 7.772311411692324e-06,
|
400 |
+
"loss": 0.2011,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.8715596330275229,
|
405 |
+
"grad_norm": 19.925289154052734,
|
406 |
+
"learning_rate": 7.743631295782021e-06,
|
407 |
+
"loss": 0.3707,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.8868501529051988,
|
412 |
+
"grad_norm": 7.974088668823242,
|
413 |
+
"learning_rate": 7.714951179871716e-06,
|
414 |
+
"loss": 0.3449,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.9021406727828746,
|
419 |
+
"grad_norm": 28.756046295166016,
|
420 |
+
"learning_rate": 7.686271063961413e-06,
|
421 |
+
"loss": 0.5487,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.9174311926605505,
|
426 |
+
"grad_norm": 2.515625238418579,
|
427 |
+
"learning_rate": 7.657590948051108e-06,
|
428 |
+
"loss": 0.3541,
|
429 |
+
"step": 600
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.9327217125382263,
|
433 |
+
"grad_norm": 7.654485702514648,
|
434 |
+
"learning_rate": 7.628910832140805e-06,
|
435 |
+
"loss": 0.3335,
|
436 |
+
"step": 610
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.9480122324159022,
|
440 |
+
"grad_norm": 17.905406951904297,
|
441 |
+
"learning_rate": 7.600230716230501e-06,
|
442 |
+
"loss": 0.3162,
|
443 |
+
"step": 620
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.963302752293578,
|
447 |
+
"grad_norm": 16.23491668701172,
|
448 |
+
"learning_rate": 7.571550600320198e-06,
|
449 |
+
"loss": 0.3626,
|
450 |
+
"step": 630
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.9785932721712538,
|
454 |
+
"grad_norm": 0.26539096236228943,
|
455 |
+
"learning_rate": 7.542870484409894e-06,
|
456 |
+
"loss": 0.2231,
|
457 |
+
"step": 640
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.9938837920489296,
|
461 |
+
"grad_norm": 1.7205305099487305,
|
462 |
+
"learning_rate": 7.51419036849959e-06,
|
463 |
+
"loss": 0.3941,
|
464 |
+
"step": 650
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 1.0,
|
468 |
+
"eval_accuracy": 0.8706962509563887,
|
469 |
+
"eval_f1": 0.7540029112081513,
|
470 |
+
"eval_loss": 0.29941147565841675,
|
471 |
+
"eval_precision": 0.6906666666666667,
|
472 |
+
"eval_recall": 0.8301282051282052,
|
473 |
+
"eval_runtime": 4.8981,
|
474 |
+
"eval_samples_per_second": 266.841,
|
475 |
+
"eval_steps_per_second": 16.741,
|
476 |
+
"step": 654
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 1.0091743119266054,
|
480 |
+
"grad_norm": 13.617008209228516,
|
481 |
+
"learning_rate": 7.485510252589286e-06,
|
482 |
+
"loss": 0.3849,
|
483 |
+
"step": 660
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 1.0244648318042813,
|
487 |
+
"grad_norm": 6.546511650085449,
|
488 |
+
"learning_rate": 7.456830136678982e-06,
|
489 |
+
"loss": 0.2871,
|
490 |
+
"step": 670
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 1.039755351681957,
|
494 |
+
"grad_norm": 5.344148635864258,
|
495 |
+
"learning_rate": 7.428150020768678e-06,
|
496 |
+
"loss": 0.2358,
|
497 |
+
"step": 680
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 1.0550458715596331,
|
501 |
+
"grad_norm": 7.6064348220825195,
|
502 |
+
"learning_rate": 7.399469904858376e-06,
|
503 |
+
"loss": 0.2451,
|
504 |
+
"step": 690
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 1.070336391437309,
|
508 |
+
"grad_norm": 21.806922912597656,
|
509 |
+
"learning_rate": 7.370789788948072e-06,
|
510 |
+
"loss": 0.2107,
|
511 |
+
"step": 700
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 1.0856269113149848,
|
515 |
+
"grad_norm": 10.673025131225586,
|
516 |
+
"learning_rate": 7.342109673037768e-06,
|
517 |
+
"loss": 0.2411,
|
518 |
+
"step": 710
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 1.1009174311926606,
|
522 |
+
"grad_norm": 14.94886589050293,
|
523 |
+
"learning_rate": 7.313429557127464e-06,
|
524 |
+
"loss": 0.1456,
|
525 |
+
"step": 720
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 1.1162079510703364,
|
529 |
+
"grad_norm": 12.618221282958984,
|
530 |
+
"learning_rate": 7.28474944121716e-06,
|
531 |
+
"loss": 0.2997,
|
532 |
+
"step": 730
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 1.1314984709480123,
|
536 |
+
"grad_norm": 0.2970161736011505,
|
537 |
+
"learning_rate": 7.256069325306856e-06,
|
538 |
+
"loss": 0.2922,
|
539 |
+
"step": 740
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 1.146788990825688,
|
543 |
+
"grad_norm": 3.0866048336029053,
|
544 |
+
"learning_rate": 7.227389209396553e-06,
|
545 |
+
"loss": 0.1767,
|
546 |
+
"step": 750
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 1.162079510703364,
|
550 |
+
"grad_norm": 12.53414535522461,
|
551 |
+
"learning_rate": 7.198709093486249e-06,
|
552 |
+
"loss": 0.2948,
|
553 |
+
"step": 760
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 1.1773700305810397,
|
557 |
+
"grad_norm": 24.84918975830078,
|
558 |
+
"learning_rate": 7.170028977575945e-06,
|
559 |
+
"loss": 0.1902,
|
560 |
+
"step": 770
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 1.1926605504587156,
|
564 |
+
"grad_norm": 1.028529167175293,
|
565 |
+
"learning_rate": 7.141348861665641e-06,
|
566 |
+
"loss": 0.2246,
|
567 |
+
"step": 780
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 1.2079510703363914,
|
571 |
+
"grad_norm": 20.015581130981445,
|
572 |
+
"learning_rate": 7.112668745755337e-06,
|
573 |
+
"loss": 0.2278,
|
574 |
+
"step": 790
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 1.2232415902140672,
|
578 |
+
"grad_norm": 12.658394813537598,
|
579 |
+
"learning_rate": 7.083988629845033e-06,
|
580 |
+
"loss": 0.3021,
|
581 |
+
"step": 800
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 1.238532110091743,
|
585 |
+
"grad_norm": 4.727680683135986,
|
586 |
+
"learning_rate": 7.05530851393473e-06,
|
587 |
+
"loss": 0.3192,
|
588 |
+
"step": 810
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 1.2538226299694188,
|
592 |
+
"grad_norm": 16.0307559967041,
|
593 |
+
"learning_rate": 7.026628398024426e-06,
|
594 |
+
"loss": 0.3534,
|
595 |
+
"step": 820
|
596 |
+
},
|
597 |
+
{
|
598 |
+
"epoch": 1.2691131498470947,
|
599 |
+
"grad_norm": 4.940106391906738,
|
600 |
+
"learning_rate": 6.997948282114122e-06,
|
601 |
+
"loss": 0.3564,
|
602 |
+
"step": 830
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 1.2844036697247707,
|
606 |
+
"grad_norm": 23.759117126464844,
|
607 |
+
"learning_rate": 6.969268166203818e-06,
|
608 |
+
"loss": 0.2746,
|
609 |
+
"step": 840
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 1.2996941896024465,
|
613 |
+
"grad_norm": 25.81449317932129,
|
614 |
+
"learning_rate": 6.9405880502935144e-06,
|
615 |
+
"loss": 0.4967,
|
616 |
+
"step": 850
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 1.3149847094801224,
|
620 |
+
"grad_norm": 4.882550239562988,
|
621 |
+
"learning_rate": 6.9119079343832104e-06,
|
622 |
+
"loss": 0.2423,
|
623 |
+
"step": 860
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 1.3302752293577982,
|
627 |
+
"grad_norm": 12.674756050109863,
|
628 |
+
"learning_rate": 6.883227818472907e-06,
|
629 |
+
"loss": 0.3532,
|
630 |
+
"step": 870
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 1.345565749235474,
|
634 |
+
"grad_norm": 13.883800506591797,
|
635 |
+
"learning_rate": 6.854547702562603e-06,
|
636 |
+
"loss": 0.2396,
|
637 |
+
"step": 880
|
638 |
+
},
|
639 |
+
{
|
640 |
+
"epoch": 1.3608562691131498,
|
641 |
+
"grad_norm": 7.71323299407959,
|
642 |
+
"learning_rate": 6.825867586652299e-06,
|
643 |
+
"loss": 0.2591,
|
644 |
+
"step": 890
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 1.3761467889908257,
|
648 |
+
"grad_norm": 2.174020767211914,
|
649 |
+
"learning_rate": 6.797187470741995e-06,
|
650 |
+
"loss": 0.2171,
|
651 |
+
"step": 900
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 1.3914373088685015,
|
655 |
+
"grad_norm": 2.351623773574829,
|
656 |
+
"learning_rate": 6.768507354831691e-06,
|
657 |
+
"loss": 0.3057,
|
658 |
+
"step": 910
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 1.4067278287461773,
|
662 |
+
"grad_norm": 26.668006896972656,
|
663 |
+
"learning_rate": 6.739827238921387e-06,
|
664 |
+
"loss": 0.2594,
|
665 |
+
"step": 920
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 1.4220183486238533,
|
669 |
+
"grad_norm": 13.497000694274902,
|
670 |
+
"learning_rate": 6.711147123011084e-06,
|
671 |
+
"loss": 0.2886,
|
672 |
+
"step": 930
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 1.4373088685015292,
|
676 |
+
"grad_norm": 9.161819458007812,
|
677 |
+
"learning_rate": 6.682467007100781e-06,
|
678 |
+
"loss": 0.2983,
|
679 |
+
"step": 940
|
680 |
+
},
|
681 |
+
{
|
682 |
+
"epoch": 1.452599388379205,
|
683 |
+
"grad_norm": 22.162334442138672,
|
684 |
+
"learning_rate": 6.653786891190477e-06,
|
685 |
+
"loss": 0.1644,
|
686 |
+
"step": 950
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 1.4678899082568808,
|
690 |
+
"grad_norm": 0.5806663632392883,
|
691 |
+
"learning_rate": 6.625106775280173e-06,
|
692 |
+
"loss": 0.2643,
|
693 |
+
"step": 960
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 1.4831804281345566,
|
697 |
+
"grad_norm": 0.31279271841049194,
|
698 |
+
"learning_rate": 6.596426659369869e-06,
|
699 |
+
"loss": 0.2888,
|
700 |
+
"step": 970
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 1.4984709480122325,
|
704 |
+
"grad_norm": 0.543879508972168,
|
705 |
+
"learning_rate": 6.567746543459565e-06,
|
706 |
+
"loss": 0.2786,
|
707 |
+
"step": 980
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 1.5137614678899083,
|
711 |
+
"grad_norm": 0.9113799333572388,
|
712 |
+
"learning_rate": 6.539066427549262e-06,
|
713 |
+
"loss": 0.4815,
|
714 |
+
"step": 990
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 1.529051987767584,
|
718 |
+
"grad_norm": 17.983909606933594,
|
719 |
+
"learning_rate": 6.510386311638958e-06,
|
720 |
+
"loss": 0.2022,
|
721 |
+
"step": 1000
|
722 |
+
},
|
723 |
+
{
|
724 |
+
"epoch": 1.54434250764526,
|
725 |
+
"grad_norm": 3.3928322792053223,
|
726 |
+
"learning_rate": 6.481706195728654e-06,
|
727 |
+
"loss": 0.2222,
|
728 |
+
"step": 1010
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 1.5596330275229358,
|
732 |
+
"grad_norm": 27.15311050415039,
|
733 |
+
"learning_rate": 6.45302607981835e-06,
|
734 |
+
"loss": 0.2879,
|
735 |
+
"step": 1020
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 1.5749235474006116,
|
739 |
+
"grad_norm": 1.4941473007202148,
|
740 |
+
"learning_rate": 6.424345963908046e-06,
|
741 |
+
"loss": 0.3431,
|
742 |
+
"step": 1030
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 1.5902140672782874,
|
746 |
+
"grad_norm": 26.38034439086914,
|
747 |
+
"learning_rate": 6.395665847997742e-06,
|
748 |
+
"loss": 0.2812,
|
749 |
+
"step": 1040
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 1.6055045871559632,
|
753 |
+
"grad_norm": 4.923500061035156,
|
754 |
+
"learning_rate": 6.366985732087439e-06,
|
755 |
+
"loss": 0.2865,
|
756 |
+
"step": 1050
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 1.620795107033639,
|
760 |
+
"grad_norm": 0.2041570544242859,
|
761 |
+
"learning_rate": 6.338305616177135e-06,
|
762 |
+
"loss": 0.4086,
|
763 |
+
"step": 1060
|
764 |
+
},
|
765 |
+
{
|
766 |
+
"epoch": 1.6360856269113149,
|
767 |
+
"grad_norm": 1.037766456604004,
|
768 |
+
"learning_rate": 6.309625500266832e-06,
|
769 |
+
"loss": 0.1451,
|
770 |
+
"step": 1070
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 1.6513761467889907,
|
774 |
+
"grad_norm": 9.758125305175781,
|
775 |
+
"learning_rate": 6.280945384356528e-06,
|
776 |
+
"loss": 0.3565,
|
777 |
+
"step": 1080
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 1.6666666666666665,
|
781 |
+
"grad_norm": 0.7395672798156738,
|
782 |
+
"learning_rate": 6.252265268446224e-06,
|
783 |
+
"loss": 0.3725,
|
784 |
+
"step": 1090
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 1.6819571865443423,
|
788 |
+
"grad_norm": 2.0258843898773193,
|
789 |
+
"learning_rate": 6.223585152535921e-06,
|
790 |
+
"loss": 0.1929,
|
791 |
+
"step": 1100
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 1.6972477064220184,
|
795 |
+
"grad_norm": 6.11789083480835,
|
796 |
+
"learning_rate": 6.194905036625617e-06,
|
797 |
+
"loss": 0.3753,
|
798 |
+
"step": 1110
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 1.7125382262996942,
|
802 |
+
"grad_norm": 11.606084823608398,
|
803 |
+
"learning_rate": 6.166224920715313e-06,
|
804 |
+
"loss": 0.227,
|
805 |
+
"step": 1120
|
806 |
+
},
|
807 |
+
{
|
808 |
+
"epoch": 1.72782874617737,
|
809 |
+
"grad_norm": 17.951644897460938,
|
810 |
+
"learning_rate": 6.137544804805009e-06,
|
811 |
+
"loss": 0.4915,
|
812 |
+
"step": 1130
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 1.7431192660550459,
|
816 |
+
"grad_norm": 0.44653165340423584,
|
817 |
+
"learning_rate": 6.108864688894705e-06,
|
818 |
+
"loss": 0.2273,
|
819 |
+
"step": 1140
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 1.7584097859327217,
|
823 |
+
"grad_norm": 17.01247215270996,
|
824 |
+
"learning_rate": 6.080184572984401e-06,
|
825 |
+
"loss": 0.3209,
|
826 |
+
"step": 1150
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 1.7737003058103975,
|
830 |
+
"grad_norm": 20.699323654174805,
|
831 |
+
"learning_rate": 6.0515044570740976e-06,
|
832 |
+
"loss": 0.435,
|
833 |
+
"step": 1160
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 1.7889908256880735,
|
837 |
+
"grad_norm": 30.34465789794922,
|
838 |
+
"learning_rate": 6.0228243411637936e-06,
|
839 |
+
"loss": 0.2472,
|
840 |
+
"step": 1170
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 1.8042813455657494,
|
844 |
+
"grad_norm": 4.261192798614502,
|
845 |
+
"learning_rate": 5.99414422525349e-06,
|
846 |
+
"loss": 0.285,
|
847 |
+
"step": 1180
|
848 |
+
},
|
849 |
+
{
|
850 |
+
"epoch": 1.8195718654434252,
|
851 |
+
"grad_norm": 20.993389129638672,
|
852 |
+
"learning_rate": 5.965464109343186e-06,
|
853 |
+
"loss": 0.3737,
|
854 |
+
"step": 1190
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.834862385321101,
|
858 |
+
"grad_norm": 0.8759846091270447,
|
859 |
+
"learning_rate": 5.936783993432882e-06,
|
860 |
+
"loss": 0.1866,
|
861 |
+
"step": 1200
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 1.8501529051987768,
|
865 |
+
"grad_norm": 29.079099655151367,
|
866 |
+
"learning_rate": 5.9081038775225785e-06,
|
867 |
+
"loss": 0.4163,
|
868 |
+
"step": 1210
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 1.8654434250764527,
|
872 |
+
"grad_norm": 3.231898546218872,
|
873 |
+
"learning_rate": 5.879423761612275e-06,
|
874 |
+
"loss": 0.4359,
|
875 |
+
"step": 1220
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 1.8807339449541285,
|
879 |
+
"grad_norm": 17.104846954345703,
|
880 |
+
"learning_rate": 5.850743645701971e-06,
|
881 |
+
"loss": 0.2606,
|
882 |
+
"step": 1230
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 1.8960244648318043,
|
886 |
+
"grad_norm": 0.409883588552475,
|
887 |
+
"learning_rate": 5.822063529791667e-06,
|
888 |
+
"loss": 0.1581,
|
889 |
+
"step": 1240
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.9113149847094801,
|
893 |
+
"grad_norm": 10.206957817077637,
|
894 |
+
"learning_rate": 5.793383413881363e-06,
|
895 |
+
"loss": 0.2835,
|
896 |
+
"step": 1250
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.926605504587156,
|
900 |
+
"grad_norm": 16.08922576904297,
|
901 |
+
"learning_rate": 5.764703297971059e-06,
|
902 |
+
"loss": 0.2338,
|
903 |
+
"step": 1260
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 1.9418960244648318,
|
907 |
+
"grad_norm": 3.5700275897979736,
|
908 |
+
"learning_rate": 5.736023182060755e-06,
|
909 |
+
"loss": 0.4643,
|
910 |
+
"step": 1270
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 1.9571865443425076,
|
914 |
+
"grad_norm": 13.477959632873535,
|
915 |
+
"learning_rate": 5.707343066150452e-06,
|
916 |
+
"loss": 0.2526,
|
917 |
+
"step": 1280
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 1.9724770642201834,
|
921 |
+
"grad_norm": 8.30550479888916,
|
922 |
+
"learning_rate": 5.678662950240148e-06,
|
923 |
+
"loss": 0.4456,
|
924 |
+
"step": 1290
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 1.9877675840978593,
|
928 |
+
"grad_norm": 7.599588871002197,
|
929 |
+
"learning_rate": 5.649982834329844e-06,
|
930 |
+
"loss": 0.259,
|
931 |
+
"step": 1300
|
932 |
+
},
|
933 |
+
{
|
934 |
+
"epoch": 2.0,
|
935 |
+
"eval_accuracy": 0.881407804131599,
|
936 |
+
"eval_f1": 0.7589424572317263,
|
937 |
+
"eval_loss": 0.3265691101551056,
|
938 |
+
"eval_precision": 0.7371601208459214,
|
939 |
+
"eval_recall": 0.782051282051282,
|
940 |
+
"eval_runtime": 4.9078,
|
941 |
+
"eval_samples_per_second": 266.31,
|
942 |
+
"eval_steps_per_second": 16.708,
|
943 |
+
"step": 1308
|
944 |
+
},
|
945 |
+
{
|
946 |
+
"epoch": 2.003058103975535,
|
947 |
+
"grad_norm": 5.8759026527404785,
|
948 |
+
"learning_rate": 5.62130271841954e-06,
|
949 |
+
"loss": 0.2964,
|
950 |
+
"step": 1310
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 2.018348623853211,
|
954 |
+
"grad_norm": 5.663475036621094,
|
955 |
+
"learning_rate": 5.592622602509236e-06,
|
956 |
+
"loss": 0.2355,
|
957 |
+
"step": 1320
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 2.0336391437308867,
|
961 |
+
"grad_norm": 12.831345558166504,
|
962 |
+
"learning_rate": 5.563942486598932e-06,
|
963 |
+
"loss": 0.2225,
|
964 |
+
"step": 1330
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 2.0489296636085625,
|
968 |
+
"grad_norm": 0.6970747709274292,
|
969 |
+
"learning_rate": 5.53526237068863e-06,
|
970 |
+
"loss": 0.279,
|
971 |
+
"step": 1340
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 2.0642201834862384,
|
975 |
+
"grad_norm": 20.61145782470703,
|
976 |
+
"learning_rate": 5.506582254778326e-06,
|
977 |
+
"loss": 0.2825,
|
978 |
+
"step": 1350
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 2.079510703363914,
|
982 |
+
"grad_norm": 4.7170891761779785,
|
983 |
+
"learning_rate": 5.477902138868022e-06,
|
984 |
+
"loss": 0.2996,
|
985 |
+
"step": 1360
|
986 |
+
},
|
987 |
+
{
|
988 |
+
"epoch": 2.09480122324159,
|
989 |
+
"grad_norm": 8.943619728088379,
|
990 |
+
"learning_rate": 5.449222022957718e-06,
|
991 |
+
"loss": 0.1539,
|
992 |
+
"step": 1370
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 2.1100917431192663,
|
996 |
+
"grad_norm": 0.7099681496620178,
|
997 |
+
"learning_rate": 5.420541907047414e-06,
|
998 |
+
"loss": 0.1958,
|
999 |
+
"step": 1380
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 2.1253822629969417,
|
1003 |
+
"grad_norm": 3.903538227081299,
|
1004 |
+
"learning_rate": 5.39186179113711e-06,
|
1005 |
+
"loss": 0.3434,
|
1006 |
+
"step": 1390
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 2.140672782874618,
|
1010 |
+
"grad_norm": 4.8174920082092285,
|
1011 |
+
"learning_rate": 5.363181675226807e-06,
|
1012 |
+
"loss": 0.266,
|
1013 |
+
"step": 1400
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 2.1559633027522938,
|
1017 |
+
"grad_norm": 13.861398696899414,
|
1018 |
+
"learning_rate": 5.334501559316503e-06,
|
1019 |
+
"loss": 0.226,
|
1020 |
+
"step": 1410
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 2.1712538226299696,
|
1024 |
+
"grad_norm": 0.26603391766548157,
|
1025 |
+
"learning_rate": 5.305821443406199e-06,
|
1026 |
+
"loss": 0.1668,
|
1027 |
+
"step": 1420
|
1028 |
+
},
|
1029 |
+
{
|
1030 |
+
"epoch": 2.1865443425076454,
|
1031 |
+
"grad_norm": 8.830183982849121,
|
1032 |
+
"learning_rate": 5.277141327495895e-06,
|
1033 |
+
"loss": 0.1988,
|
1034 |
+
"step": 1430
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 2.2018348623853212,
|
1038 |
+
"grad_norm": 0.7192776799201965,
|
1039 |
+
"learning_rate": 5.248461211585591e-06,
|
1040 |
+
"loss": 0.2137,
|
1041 |
+
"step": 1440
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 2.217125382262997,
|
1045 |
+
"grad_norm": 24.46476173400879,
|
1046 |
+
"learning_rate": 5.219781095675287e-06,
|
1047 |
+
"loss": 0.3146,
|
1048 |
+
"step": 1450
|
1049 |
+
},
|
1050 |
+
{
|
1051 |
+
"epoch": 2.232415902140673,
|
1052 |
+
"grad_norm": 36.542945861816406,
|
1053 |
+
"learning_rate": 5.191100979764985e-06,
|
1054 |
+
"loss": 0.3024,
|
1055 |
+
"step": 1460
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 2.2477064220183487,
|
1059 |
+
"grad_norm": 18.665542602539062,
|
1060 |
+
"learning_rate": 5.162420863854681e-06,
|
1061 |
+
"loss": 0.45,
|
1062 |
+
"step": 1470
|
1063 |
+
},
|
1064 |
+
{
|
1065 |
+
"epoch": 2.2629969418960245,
|
1066 |
+
"grad_norm": 0.5163823366165161,
|
1067 |
+
"learning_rate": 5.133740747944377e-06,
|
1068 |
+
"loss": 0.2063,
|
1069 |
+
"step": 1480
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 2.2782874617737003,
|
1073 |
+
"grad_norm": 12.623830795288086,
|
1074 |
+
"learning_rate": 5.105060632034073e-06,
|
1075 |
+
"loss": 0.3537,
|
1076 |
+
"step": 1490
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 2.293577981651376,
|
1080 |
+
"grad_norm": 6.660862922668457,
|
1081 |
+
"learning_rate": 5.076380516123769e-06,
|
1082 |
+
"loss": 0.3763,
|
1083 |
+
"step": 1500
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 2.308868501529052,
|
1087 |
+
"grad_norm": 18.302440643310547,
|
1088 |
+
"learning_rate": 5.047700400213465e-06,
|
1089 |
+
"loss": 0.2796,
|
1090 |
+
"step": 1510
|
1091 |
+
},
|
1092 |
+
{
|
1093 |
+
"epoch": 2.324159021406728,
|
1094 |
+
"grad_norm": 5.233463764190674,
|
1095 |
+
"learning_rate": 5.019020284303162e-06,
|
1096 |
+
"loss": 0.2484,
|
1097 |
+
"step": 1520
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 2.3394495412844036,
|
1101 |
+
"grad_norm": 0.2765680253505707,
|
1102 |
+
"learning_rate": 4.990340168392858e-06,
|
1103 |
+
"loss": 0.197,
|
1104 |
+
"step": 1530
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 2.3547400611620795,
|
1108 |
+
"grad_norm": 7.559242248535156,
|
1109 |
+
"learning_rate": 4.961660052482554e-06,
|
1110 |
+
"loss": 0.2913,
|
1111 |
+
"step": 1540
|
1112 |
+
},
|
1113 |
+
{
|
1114 |
+
"epoch": 2.3700305810397553,
|
1115 |
+
"grad_norm": 20.358930587768555,
|
1116 |
+
"learning_rate": 4.93297993657225e-06,
|
1117 |
+
"loss": 0.2762,
|
1118 |
+
"step": 1550
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 2.385321100917431,
|
1122 |
+
"grad_norm": 23.49895668029785,
|
1123 |
+
"learning_rate": 4.904299820661946e-06,
|
1124 |
+
"loss": 0.3327,
|
1125 |
+
"step": 1560
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 2.400611620795107,
|
1129 |
+
"grad_norm": 9.241127014160156,
|
1130 |
+
"learning_rate": 4.875619704751642e-06,
|
1131 |
+
"loss": 0.2981,
|
1132 |
+
"step": 1570
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 2.4159021406727827,
|
1136 |
+
"grad_norm": 9.627830505371094,
|
1137 |
+
"learning_rate": 4.846939588841339e-06,
|
1138 |
+
"loss": 0.3052,
|
1139 |
+
"step": 1580
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 2.4311926605504586,
|
1143 |
+
"grad_norm": 6.3946075439453125,
|
1144 |
+
"learning_rate": 4.818259472931035e-06,
|
1145 |
+
"loss": 0.1022,
|
1146 |
+
"step": 1590
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 2.4464831804281344,
|
1150 |
+
"grad_norm": 0.6927328109741211,
|
1151 |
+
"learning_rate": 4.789579357020731e-06,
|
1152 |
+
"loss": 0.0922,
|
1153 |
+
"step": 1600
|
1154 |
+
},
|
1155 |
+
{
|
1156 |
+
"epoch": 2.46177370030581,
|
1157 |
+
"grad_norm": 1.498061180114746,
|
1158 |
+
"learning_rate": 4.760899241110427e-06,
|
1159 |
+
"loss": 0.1883,
|
1160 |
+
"step": 1610
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 2.477064220183486,
|
1164 |
+
"grad_norm": 3.1200311183929443,
|
1165 |
+
"learning_rate": 4.732219125200123e-06,
|
1166 |
+
"loss": 0.1963,
|
1167 |
+
"step": 1620
|
1168 |
+
},
|
1169 |
+
{
|
1170 |
+
"epoch": 2.4923547400611623,
|
1171 |
+
"grad_norm": 5.31001615524292,
|
1172 |
+
"learning_rate": 4.703539009289819e-06,
|
1173 |
+
"loss": 0.3003,
|
1174 |
+
"step": 1630
|
1175 |
+
},
|
1176 |
+
{
|
1177 |
+
"epoch": 2.5076452599388377,
|
1178 |
+
"grad_norm": 13.968355178833008,
|
1179 |
+
"learning_rate": 4.674858893379516e-06,
|
1180 |
+
"loss": 0.3723,
|
1181 |
+
"step": 1640
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 2.522935779816514,
|
1185 |
+
"grad_norm": 30.514612197875977,
|
1186 |
+
"learning_rate": 4.646178777469212e-06,
|
1187 |
+
"loss": 0.2453,
|
1188 |
+
"step": 1650
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 2.5382262996941893,
|
1192 |
+
"grad_norm": 18.336116790771484,
|
1193 |
+
"learning_rate": 4.617498661558908e-06,
|
1194 |
+
"loss": 0.2514,
|
1195 |
+
"step": 1660
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 2.5535168195718656,
|
1199 |
+
"grad_norm": 3.3345324993133545,
|
1200 |
+
"learning_rate": 4.588818545648604e-06,
|
1201 |
+
"loss": 0.1784,
|
1202 |
+
"step": 1670
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 2.5688073394495414,
|
1206 |
+
"grad_norm": 0.6828510165214539,
|
1207 |
+
"learning_rate": 4.560138429738301e-06,
|
1208 |
+
"loss": 0.1926,
|
1209 |
+
"step": 1680
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 2.5840978593272173,
|
1213 |
+
"grad_norm": 0.3491632044315338,
|
1214 |
+
"learning_rate": 4.531458313827997e-06,
|
1215 |
+
"loss": 0.3459,
|
1216 |
+
"step": 1690
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 2.599388379204893,
|
1220 |
+
"grad_norm": 14.196633338928223,
|
1221 |
+
"learning_rate": 4.502778197917693e-06,
|
1222 |
+
"loss": 0.2798,
|
1223 |
+
"step": 1700
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 2.614678899082569,
|
1227 |
+
"grad_norm": 3.067326307296753,
|
1228 |
+
"learning_rate": 4.47409808200739e-06,
|
1229 |
+
"loss": 0.2548,
|
1230 |
+
"step": 1710
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 2.6299694189602447,
|
1234 |
+
"grad_norm": 10.058695793151855,
|
1235 |
+
"learning_rate": 4.445417966097086e-06,
|
1236 |
+
"loss": 0.3088,
|
1237 |
+
"step": 1720
|
1238 |
+
},
|
1239 |
+
{
|
1240 |
+
"epoch": 2.6452599388379205,
|
1241 |
+
"grad_norm": 24.748844146728516,
|
1242 |
+
"learning_rate": 4.416737850186782e-06,
|
1243 |
+
"loss": 0.2524,
|
1244 |
+
"step": 1730
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 2.6605504587155964,
|
1248 |
+
"grad_norm": 9.57737922668457,
|
1249 |
+
"learning_rate": 4.388057734276478e-06,
|
1250 |
+
"loss": 0.1089,
|
1251 |
+
"step": 1740
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 2.675840978593272,
|
1255 |
+
"grad_norm": 4.255125999450684,
|
1256 |
+
"learning_rate": 4.359377618366174e-06,
|
1257 |
+
"loss": 0.2619,
|
1258 |
+
"step": 1750
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 2.691131498470948,
|
1262 |
+
"grad_norm": 12.66395378112793,
|
1263 |
+
"learning_rate": 4.33069750245587e-06,
|
1264 |
+
"loss": 0.113,
|
1265 |
+
"step": 1760
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 2.706422018348624,
|
1269 |
+
"grad_norm": 27.81719970703125,
|
1270 |
+
"learning_rate": 4.302017386545567e-06,
|
1271 |
+
"loss": 0.1587,
|
1272 |
+
"step": 1770
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 2.7217125382262997,
|
1276 |
+
"grad_norm": 9.098254203796387,
|
1277 |
+
"learning_rate": 4.273337270635263e-06,
|
1278 |
+
"loss": 0.1325,
|
1279 |
+
"step": 1780
|
1280 |
+
},
|
1281 |
+
{
|
1282 |
+
"epoch": 2.7370030581039755,
|
1283 |
+
"grad_norm": 7.441411018371582,
|
1284 |
+
"learning_rate": 4.244657154724959e-06,
|
1285 |
+
"loss": 0.4097,
|
1286 |
+
"step": 1790
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 2.7522935779816513,
|
1290 |
+
"grad_norm": 28.30503273010254,
|
1291 |
+
"learning_rate": 4.215977038814656e-06,
|
1292 |
+
"loss": 0.2826,
|
1293 |
+
"step": 1800
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 2.767584097859327,
|
1297 |
+
"grad_norm": 47.75388717651367,
|
1298 |
+
"learning_rate": 4.187296922904352e-06,
|
1299 |
+
"loss": 0.2992,
|
1300 |
+
"step": 1810
|
1301 |
+
},
|
1302 |
+
{
|
1303 |
+
"epoch": 2.782874617737003,
|
1304 |
+
"grad_norm": 5.235270023345947,
|
1305 |
+
"learning_rate": 4.158616806994048e-06,
|
1306 |
+
"loss": 0.1981,
|
1307 |
+
"step": 1820
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 2.7981651376146788,
|
1311 |
+
"grad_norm": 0.08484458923339844,
|
1312 |
+
"learning_rate": 4.129936691083745e-06,
|
1313 |
+
"loss": 0.3315,
|
1314 |
+
"step": 1830
|
1315 |
+
},
|
1316 |
+
{
|
1317 |
+
"epoch": 2.8134556574923546,
|
1318 |
+
"grad_norm": 17.93990707397461,
|
1319 |
+
"learning_rate": 4.101256575173441e-06,
|
1320 |
+
"loss": 0.3326,
|
1321 |
+
"step": 1840
|
1322 |
+
},
|
1323 |
+
{
|
1324 |
+
"epoch": 2.8287461773700304,
|
1325 |
+
"grad_norm": 0.8848928213119507,
|
1326 |
+
"learning_rate": 4.072576459263137e-06,
|
1327 |
+
"loss": 0.1852,
|
1328 |
+
"step": 1850
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 2.8440366972477067,
|
1332 |
+
"grad_norm": 1.0666664838790894,
|
1333 |
+
"learning_rate": 4.043896343352833e-06,
|
1334 |
+
"loss": 0.1235,
|
1335 |
+
"step": 1860
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 2.859327217125382,
|
1339 |
+
"grad_norm": 29.053184509277344,
|
1340 |
+
"learning_rate": 4.015216227442529e-06,
|
1341 |
+
"loss": 0.309,
|
1342 |
+
"step": 1870
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 2.8746177370030583,
|
1346 |
+
"grad_norm": 30.479965209960938,
|
1347 |
+
"learning_rate": 3.986536111532225e-06,
|
1348 |
+
"loss": 0.2986,
|
1349 |
+
"step": 1880
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 2.8899082568807337,
|
1353 |
+
"grad_norm": 23.085195541381836,
|
1354 |
+
"learning_rate": 3.957855995621922e-06,
|
1355 |
+
"loss": 0.1596,
|
1356 |
+
"step": 1890
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 2.90519877675841,
|
1360 |
+
"grad_norm": 1.5042694807052612,
|
1361 |
+
"learning_rate": 3.929175879711618e-06,
|
1362 |
+
"loss": 0.2213,
|
1363 |
+
"step": 1900
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 2.9204892966360854,
|
1367 |
+
"grad_norm": 22.314062118530273,
|
1368 |
+
"learning_rate": 3.900495763801314e-06,
|
1369 |
+
"loss": 0.1854,
|
1370 |
+
"step": 1910
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 2.9357798165137616,
|
1374 |
+
"grad_norm": 3.0091068744659424,
|
1375 |
+
"learning_rate": 3.8718156478910105e-06,
|
1376 |
+
"loss": 0.1231,
|
1377 |
+
"step": 1920
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 2.9510703363914375,
|
1381 |
+
"grad_norm": 0.4003874361515045,
|
1382 |
+
"learning_rate": 3.8431355319807065e-06,
|
1383 |
+
"loss": 0.1013,
|
1384 |
+
"step": 1930
|
1385 |
+
},
|
1386 |
+
{
|
1387 |
+
"epoch": 2.9663608562691133,
|
1388 |
+
"grad_norm": 36.91960906982422,
|
1389 |
+
"learning_rate": 3.8144554160704025e-06,
|
1390 |
+
"loss": 0.1974,
|
1391 |
+
"step": 1940
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 2.981651376146789,
|
1395 |
+
"grad_norm": 10.349553108215332,
|
1396 |
+
"learning_rate": 3.785775300160099e-06,
|
1397 |
+
"loss": 0.3974,
|
1398 |
+
"step": 1950
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 2.996941896024465,
|
1402 |
+
"grad_norm": 4.602114677429199,
|
1403 |
+
"learning_rate": 3.757095184249795e-06,
|
1404 |
+
"loss": 0.1764,
|
1405 |
+
"step": 1960
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 3.0,
|
1409 |
+
"eval_accuracy": 0.8837031369548585,
|
1410 |
+
"eval_f1": 0.7661538461538462,
|
1411 |
+
"eval_loss": 0.416372686624527,
|
1412 |
+
"eval_precision": 0.7366863905325444,
|
1413 |
+
"eval_recall": 0.7980769230769231,
|
1414 |
+
"eval_runtime": 4.9368,
|
1415 |
+
"eval_samples_per_second": 264.749,
|
1416 |
+
"eval_steps_per_second": 16.61,
|
1417 |
+
"step": 1962
|
1418 |
+
},
|
1419 |
+
{
|
1420 |
+
"epoch": 3.0122324159021407,
|
1421 |
+
"grad_norm": 0.9072074890136719,
|
1422 |
+
"learning_rate": 3.728415068339491e-06,
|
1423 |
+
"loss": 0.0342,
|
1424 |
+
"step": 1970
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 3.0275229357798166,
|
1428 |
+
"grad_norm": 11.377812385559082,
|
1429 |
+
"learning_rate": 3.699734952429188e-06,
|
1430 |
+
"loss": 0.1661,
|
1431 |
+
"step": 1980
|
1432 |
+
},
|
1433 |
+
{
|
1434 |
+
"epoch": 3.0428134556574924,
|
1435 |
+
"grad_norm": 17.510143280029297,
|
1436 |
+
"learning_rate": 3.671054836518884e-06,
|
1437 |
+
"loss": 0.1267,
|
1438 |
+
"step": 1990
|
1439 |
+
},
|
1440 |
+
{
|
1441 |
+
"epoch": 3.058103975535168,
|
1442 |
+
"grad_norm": 0.12311755865812302,
|
1443 |
+
"learning_rate": 3.64237472060858e-06,
|
1444 |
+
"loss": 0.272,
|
1445 |
+
"step": 2000
|
1446 |
+
},
|
1447 |
+
{
|
1448 |
+
"epoch": 3.073394495412844,
|
1449 |
+
"grad_norm": 7.080625534057617,
|
1450 |
+
"learning_rate": 3.6136946046982763e-06,
|
1451 |
+
"loss": 0.3724,
|
1452 |
+
"step": 2010
|
1453 |
+
},
|
1454 |
+
{
|
1455 |
+
"epoch": 3.08868501529052,
|
1456 |
+
"grad_norm": 15.893166542053223,
|
1457 |
+
"learning_rate": 3.5850144887879723e-06,
|
1458 |
+
"loss": 0.211,
|
1459 |
+
"step": 2020
|
1460 |
+
},
|
1461 |
+
{
|
1462 |
+
"epoch": 3.1039755351681957,
|
1463 |
+
"grad_norm": 70.26548767089844,
|
1464 |
+
"learning_rate": 3.5563343728776683e-06,
|
1465 |
+
"loss": 0.1047,
|
1466 |
+
"step": 2030
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 3.1192660550458715,
|
1470 |
+
"grad_norm": 1.035912036895752,
|
1471 |
+
"learning_rate": 3.527654256967365e-06,
|
1472 |
+
"loss": 0.0586,
|
1473 |
+
"step": 2040
|
1474 |
+
},
|
1475 |
+
{
|
1476 |
+
"epoch": 3.1345565749235473,
|
1477 |
+
"grad_norm": 21.898548126220703,
|
1478 |
+
"learning_rate": 3.498974141057061e-06,
|
1479 |
+
"loss": 0.2289,
|
1480 |
+
"step": 2050
|
1481 |
+
},
|
1482 |
+
{
|
1483 |
+
"epoch": 3.149847094801223,
|
1484 |
+
"grad_norm": 9.897711753845215,
|
1485 |
+
"learning_rate": 3.4702940251467572e-06,
|
1486 |
+
"loss": 0.2056,
|
1487 |
+
"step": 2060
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 3.165137614678899,
|
1491 |
+
"grad_norm": 0.49520930647850037,
|
1492 |
+
"learning_rate": 3.4416139092364537e-06,
|
1493 |
+
"loss": 0.2293,
|
1494 |
+
"step": 2070
|
1495 |
+
},
|
1496 |
+
{
|
1497 |
+
"epoch": 3.180428134556575,
|
1498 |
+
"grad_norm": 0.14918214082717896,
|
1499 |
+
"learning_rate": 3.4129337933261497e-06,
|
1500 |
+
"loss": 0.1414,
|
1501 |
+
"step": 2080
|
1502 |
+
},
|
1503 |
+
{
|
1504 |
+
"epoch": 3.1957186544342506,
|
1505 |
+
"grad_norm": 0.4814736247062683,
|
1506 |
+
"learning_rate": 3.3842536774158457e-06,
|
1507 |
+
"loss": 0.2611,
|
1508 |
+
"step": 2090
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 3.2110091743119265,
|
1512 |
+
"grad_norm": 0.19969391822814941,
|
1513 |
+
"learning_rate": 3.355573561505542e-06,
|
1514 |
+
"loss": 0.1482,
|
1515 |
+
"step": 2100
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 3.2262996941896023,
|
1519 |
+
"grad_norm": 0.16247302293777466,
|
1520 |
+
"learning_rate": 3.3268934455952385e-06,
|
1521 |
+
"loss": 0.4237,
|
1522 |
+
"step": 2110
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 3.241590214067278,
|
1526 |
+
"grad_norm": 17.90416717529297,
|
1527 |
+
"learning_rate": 3.2982133296849346e-06,
|
1528 |
+
"loss": 0.127,
|
1529 |
+
"step": 2120
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 3.2568807339449544,
|
1533 |
+
"grad_norm": 34.14628219604492,
|
1534 |
+
"learning_rate": 3.269533213774631e-06,
|
1535 |
+
"loss": 0.1481,
|
1536 |
+
"step": 2130
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 3.2721712538226297,
|
1540 |
+
"grad_norm": 0.18314845860004425,
|
1541 |
+
"learning_rate": 3.240853097864327e-06,
|
1542 |
+
"loss": 0.1436,
|
1543 |
+
"step": 2140
|
1544 |
+
},
|
1545 |
+
{
|
1546 |
+
"epoch": 3.287461773700306,
|
1547 |
+
"grad_norm": 0.13257867097854614,
|
1548 |
+
"learning_rate": 3.212172981954023e-06,
|
1549 |
+
"loss": 0.1203,
|
1550 |
+
"step": 2150
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 3.302752293577982,
|
1554 |
+
"grad_norm": 42.67006301879883,
|
1555 |
+
"learning_rate": 3.1834928660437194e-06,
|
1556 |
+
"loss": 0.2017,
|
1557 |
+
"step": 2160
|
1558 |
+
},
|
1559 |
+
{
|
1560 |
+
"epoch": 3.3180428134556577,
|
1561 |
+
"grad_norm": 0.3060612082481384,
|
1562 |
+
"learning_rate": 3.154812750133416e-06,
|
1563 |
+
"loss": 0.1398,
|
1564 |
+
"step": 2170
|
1565 |
+
},
|
1566 |
+
{
|
1567 |
+
"epoch": 3.3333333333333335,
|
1568 |
+
"grad_norm": 6.72953987121582,
|
1569 |
+
"learning_rate": 3.126132634223112e-06,
|
1570 |
+
"loss": 0.1634,
|
1571 |
+
"step": 2180
|
1572 |
+
},
|
1573 |
+
{
|
1574 |
+
"epoch": 3.3486238532110093,
|
1575 |
+
"grad_norm": 0.4008305072784424,
|
1576 |
+
"learning_rate": 3.0974525183128083e-06,
|
1577 |
+
"loss": 0.0838,
|
1578 |
+
"step": 2190
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 3.363914373088685,
|
1582 |
+
"grad_norm": 0.2475535124540329,
|
1583 |
+
"learning_rate": 3.0687724024025043e-06,
|
1584 |
+
"loss": 0.2601,
|
1585 |
+
"step": 2200
|
1586 |
+
},
|
1587 |
+
{
|
1588 |
+
"epoch": 3.379204892966361,
|
1589 |
+
"grad_norm": 0.6882356405258179,
|
1590 |
+
"learning_rate": 3.0400922864922004e-06,
|
1591 |
+
"loss": 0.2079,
|
1592 |
+
"step": 2210
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 3.3944954128440368,
|
1596 |
+
"grad_norm": 0.11873319745063782,
|
1597 |
+
"learning_rate": 3.0114121705818968e-06,
|
1598 |
+
"loss": 0.1479,
|
1599 |
+
"step": 2220
|
1600 |
+
},
|
1601 |
+
{
|
1602 |
+
"epoch": 3.4097859327217126,
|
1603 |
+
"grad_norm": 2.4591195583343506,
|
1604 |
+
"learning_rate": 2.982732054671593e-06,
|
1605 |
+
"loss": 0.0835,
|
1606 |
+
"step": 2230
|
1607 |
+
},
|
1608 |
+
{
|
1609 |
+
"epoch": 3.4250764525993884,
|
1610 |
+
"grad_norm": 33.77700424194336,
|
1611 |
+
"learning_rate": 2.9540519387612892e-06,
|
1612 |
+
"loss": 0.325,
|
1613 |
+
"step": 2240
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 3.4403669724770642,
|
1617 |
+
"grad_norm": 0.19437803328037262,
|
1618 |
+
"learning_rate": 2.9253718228509857e-06,
|
1619 |
+
"loss": 0.147,
|
1620 |
+
"step": 2250
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 3.45565749235474,
|
1624 |
+
"grad_norm": 5.968377113342285,
|
1625 |
+
"learning_rate": 2.8966917069406817e-06,
|
1626 |
+
"loss": 0.3554,
|
1627 |
+
"step": 2260
|
1628 |
+
},
|
1629 |
+
{
|
1630 |
+
"epoch": 3.470948012232416,
|
1631 |
+
"grad_norm": 12.378327369689941,
|
1632 |
+
"learning_rate": 2.8680115910303777e-06,
|
1633 |
+
"loss": 0.2007,
|
1634 |
+
"step": 2270
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 3.4862385321100917,
|
1638 |
+
"grad_norm": 0.21296323835849762,
|
1639 |
+
"learning_rate": 2.839331475120074e-06,
|
1640 |
+
"loss": 0.2296,
|
1641 |
+
"step": 2280
|
1642 |
+
},
|
1643 |
+
{
|
1644 |
+
"epoch": 3.5015290519877675,
|
1645 |
+
"grad_norm": 13.099735260009766,
|
1646 |
+
"learning_rate": 2.81065135920977e-06,
|
1647 |
+
"loss": 0.1264,
|
1648 |
+
"step": 2290
|
1649 |
+
},
|
1650 |
+
{
|
1651 |
+
"epoch": 3.5168195718654434,
|
1652 |
+
"grad_norm": 0.31538063287734985,
|
1653 |
+
"learning_rate": 2.781971243299466e-06,
|
1654 |
+
"loss": 0.4666,
|
1655 |
+
"step": 2300
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 3.532110091743119,
|
1659 |
+
"grad_norm": 12.299838066101074,
|
1660 |
+
"learning_rate": 2.753291127389163e-06,
|
1661 |
+
"loss": 0.2505,
|
1662 |
+
"step": 2310
|
1663 |
+
},
|
1664 |
+
{
|
1665 |
+
"epoch": 3.547400611620795,
|
1666 |
+
"grad_norm": 16.644168853759766,
|
1667 |
+
"learning_rate": 2.724611011478859e-06,
|
1668 |
+
"loss": 0.4427,
|
1669 |
+
"step": 2320
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 3.562691131498471,
|
1673 |
+
"grad_norm": 39.950279235839844,
|
1674 |
+
"learning_rate": 2.695930895568555e-06,
|
1675 |
+
"loss": 0.3459,
|
1676 |
+
"step": 2330
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 3.5779816513761467,
|
1680 |
+
"grad_norm": 5.1495256423950195,
|
1681 |
+
"learning_rate": 2.6672507796582515e-06,
|
1682 |
+
"loss": 0.3096,
|
1683 |
+
"step": 2340
|
1684 |
+
},
|
1685 |
+
{
|
1686 |
+
"epoch": 3.5932721712538225,
|
1687 |
+
"grad_norm": 35.178226470947266,
|
1688 |
+
"learning_rate": 2.6385706637479475e-06,
|
1689 |
+
"loss": 0.2467,
|
1690 |
+
"step": 2350
|
1691 |
+
},
|
1692 |
+
{
|
1693 |
+
"epoch": 3.6085626911314987,
|
1694 |
+
"grad_norm": 6.108614921569824,
|
1695 |
+
"learning_rate": 2.6098905478376435e-06,
|
1696 |
+
"loss": 0.0156,
|
1697 |
+
"step": 2360
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 3.623853211009174,
|
1701 |
+
"grad_norm": 0.34197351336479187,
|
1702 |
+
"learning_rate": 2.5812104319273403e-06,
|
1703 |
+
"loss": 0.1566,
|
1704 |
+
"step": 2370
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 3.6391437308868504,
|
1708 |
+
"grad_norm": 5.665993690490723,
|
1709 |
+
"learning_rate": 2.5525303160170364e-06,
|
1710 |
+
"loss": 0.268,
|
1711 |
+
"step": 2380
|
1712 |
+
},
|
1713 |
+
{
|
1714 |
+
"epoch": 3.6544342507645258,
|
1715 |
+
"grad_norm": 3.0416011810302734,
|
1716 |
+
"learning_rate": 2.5238502001067324e-06,
|
1717 |
+
"loss": 0.1028,
|
1718 |
+
"step": 2390
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 3.669724770642202,
|
1722 |
+
"grad_norm": 5.079551696777344,
|
1723 |
+
"learning_rate": 2.495170084196429e-06,
|
1724 |
+
"loss": 0.2225,
|
1725 |
+
"step": 2400
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 3.6850152905198774,
|
1729 |
+
"grad_norm": 0.19564682245254517,
|
1730 |
+
"learning_rate": 2.466489968286125e-06,
|
1731 |
+
"loss": 0.1261,
|
1732 |
+
"step": 2410
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 3.7003058103975537,
|
1736 |
+
"grad_norm": 15.542204856872559,
|
1737 |
+
"learning_rate": 2.437809852375821e-06,
|
1738 |
+
"loss": 0.1688,
|
1739 |
+
"step": 2420
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 3.7155963302752295,
|
1743 |
+
"grad_norm": 0.636468231678009,
|
1744 |
+
"learning_rate": 2.4091297364655177e-06,
|
1745 |
+
"loss": 0.3542,
|
1746 |
+
"step": 2430
|
1747 |
+
},
|
1748 |
+
{
|
1749 |
+
"epoch": 3.7308868501529053,
|
1750 |
+
"grad_norm": 0.26796048879623413,
|
1751 |
+
"learning_rate": 2.3804496205552137e-06,
|
1752 |
+
"loss": 0.0689,
|
1753 |
+
"step": 2440
|
1754 |
+
},
|
1755 |
+
{
|
1756 |
+
"epoch": 3.746177370030581,
|
1757 |
+
"grad_norm": 9.50084400177002,
|
1758 |
+
"learning_rate": 2.3517695046449097e-06,
|
1759 |
+
"loss": 0.3902,
|
1760 |
+
"step": 2450
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 3.761467889908257,
|
1764 |
+
"grad_norm": 1.7848141193389893,
|
1765 |
+
"learning_rate": 2.323089388734606e-06,
|
1766 |
+
"loss": 0.0613,
|
1767 |
+
"step": 2460
|
1768 |
+
},
|
1769 |
+
{
|
1770 |
+
"epoch": 3.776758409785933,
|
1771 |
+
"grad_norm": 20.41553497314453,
|
1772 |
+
"learning_rate": 2.294409272824302e-06,
|
1773 |
+
"loss": 0.2541,
|
1774 |
+
"step": 2470
|
1775 |
+
},
|
1776 |
+
{
|
1777 |
+
"epoch": 3.7920489296636086,
|
1778 |
+
"grad_norm": 1.6424323320388794,
|
1779 |
+
"learning_rate": 2.2657291569139986e-06,
|
1780 |
+
"loss": 0.2198,
|
1781 |
+
"step": 2480
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 3.8073394495412844,
|
1785 |
+
"grad_norm": 0.24343428015708923,
|
1786 |
+
"learning_rate": 2.237049041003695e-06,
|
1787 |
+
"loss": 0.2029,
|
1788 |
+
"step": 2490
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 3.8226299694189603,
|
1792 |
+
"grad_norm": 44.71451950073242,
|
1793 |
+
"learning_rate": 2.208368925093391e-06,
|
1794 |
+
"loss": 0.3573,
|
1795 |
+
"step": 2500
|
1796 |
+
},
|
1797 |
+
{
|
1798 |
+
"epoch": 3.837920489296636,
|
1799 |
+
"grad_norm": 30.519176483154297,
|
1800 |
+
"learning_rate": 2.179688809183087e-06,
|
1801 |
+
"loss": 0.3304,
|
1802 |
+
"step": 2510
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 3.853211009174312,
|
1806 |
+
"grad_norm": 25.292348861694336,
|
1807 |
+
"learning_rate": 2.1510086932727835e-06,
|
1808 |
+
"loss": 0.1079,
|
1809 |
+
"step": 2520
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 3.8685015290519877,
|
1813 |
+
"grad_norm": 35.308006286621094,
|
1814 |
+
"learning_rate": 2.1223285773624795e-06,
|
1815 |
+
"loss": 0.1774,
|
1816 |
+
"step": 2530
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 3.8837920489296636,
|
1820 |
+
"grad_norm": 0.16474059224128723,
|
1821 |
+
"learning_rate": 2.093648461452176e-06,
|
1822 |
+
"loss": 0.0328,
|
1823 |
+
"step": 2540
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 3.8990825688073394,
|
1827 |
+
"grad_norm": 0.09989167749881744,
|
1828 |
+
"learning_rate": 2.0649683455418724e-06,
|
1829 |
+
"loss": 0.1969,
|
1830 |
+
"step": 2550
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 3.914373088685015,
|
1834 |
+
"grad_norm": 18.91350746154785,
|
1835 |
+
"learning_rate": 2.0362882296315684e-06,
|
1836 |
+
"loss": 0.1445,
|
1837 |
+
"step": 2560
|
1838 |
+
},
|
1839 |
+
{
|
1840 |
+
"epoch": 3.929663608562691,
|
1841 |
+
"grad_norm": 0.08901531994342804,
|
1842 |
+
"learning_rate": 2.0076081137212644e-06,
|
1843 |
+
"loss": 0.3773,
|
1844 |
+
"step": 2570
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 3.944954128440367,
|
1848 |
+
"grad_norm": 46.0722770690918,
|
1849 |
+
"learning_rate": 1.978927997810961e-06,
|
1850 |
+
"loss": 0.3817,
|
1851 |
+
"step": 2580
|
1852 |
+
},
|
1853 |
+
{
|
1854 |
+
"epoch": 3.9602446483180427,
|
1855 |
+
"grad_norm": 7.898420810699463,
|
1856 |
+
"learning_rate": 1.950247881900657e-06,
|
1857 |
+
"loss": 0.1688,
|
1858 |
+
"step": 2590
|
1859 |
+
},
|
1860 |
+
{
|
1861 |
+
"epoch": 3.9755351681957185,
|
1862 |
+
"grad_norm": 51.515220642089844,
|
1863 |
+
"learning_rate": 1.9215677659903533e-06,
|
1864 |
+
"loss": 0.1521,
|
1865 |
+
"step": 2600
|
1866 |
+
},
|
1867 |
+
{
|
1868 |
+
"epoch": 3.9908256880733948,
|
1869 |
+
"grad_norm": 0.9271498918533325,
|
1870 |
+
"learning_rate": 1.8928876500800495e-06,
|
1871 |
+
"loss": 0.1549,
|
1872 |
+
"step": 2610
|
1873 |
+
},
|
1874 |
+
{
|
1875 |
+
"epoch": 4.0,
|
1876 |
+
"eval_accuracy": 0.8859984697781178,
|
1877 |
+
"eval_f1": 0.7585089141004863,
|
1878 |
+
"eval_loss": 0.44139671325683594,
|
1879 |
+
"eval_precision": 0.7672131147540984,
|
1880 |
+
"eval_recall": 0.75,
|
1881 |
+
"eval_runtime": 4.9043,
|
1882 |
+
"eval_samples_per_second": 266.502,
|
1883 |
+
"eval_steps_per_second": 16.72,
|
1884 |
+
"step": 2616
|
1885 |
+
}
|
1886 |
+
],
|
1887 |
+
"logging_steps": 10,
|
1888 |
+
"max_steps": 3270,
|
1889 |
+
"num_input_tokens_seen": 0,
|
1890 |
+
"num_train_epochs": 5,
|
1891 |
+
"save_steps": 500,
|
1892 |
+
"stateful_callbacks": {
|
1893 |
+
"TrainerControl": {
|
1894 |
+
"args": {
|
1895 |
+
"should_epoch_stop": false,
|
1896 |
+
"should_evaluate": false,
|
1897 |
+
"should_log": false,
|
1898 |
+
"should_save": true,
|
1899 |
+
"should_training_stop": false
|
1900 |
+
},
|
1901 |
+
"attributes": {}
|
1902 |
+
}
|
1903 |
+
},
|
1904 |
+
"total_flos": 2764329675153408.0,
|
1905 |
+
"train_batch_size": 8,
|
1906 |
+
"trial_name": null,
|
1907 |
+
"trial_params": {
|
1908 |
+
"learning_rate": 9.378397902669336e-06,
|
1909 |
+
"num_train_epochs": 5,
|
1910 |
+
"per_device_train_batch_size": 8,
|
1911 |
+
"seed": 28
|
1912 |
+
}
|
1913 |
+
}
|
run-0/checkpoint-2616/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:40f24b9f959faf9f78234785503a7c7dfa403799a1c001c4311a87daac4c4bbd
|
3 |
+
size 5304
|
run-0/checkpoint-2616/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-1/checkpoint-654/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
run-1/checkpoint-654/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e405c55984cc27cd3e03d775338134281503bb3531f22447f2059c544e59f974
|
3 |
+
size 541317368
|
run-1/checkpoint-654/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4abcb6fa6739bd040bfeaa4c5f049fc4925654fa44c189bee6a289c643b29dfa
|
3 |
+
size 1082696890
|
run-1/checkpoint-654/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:38a2e74768590541c8a0877c5c1da8de71a1eebc687a1582315c96c87c8f2c4a
|
3 |
+
size 14244
|
run-1/checkpoint-654/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d1c3c746a4e9df76413db7c5ea09b20c2d96f0564f1eef2b90a35c7161b5e07
|
3 |
+
size 1064
|
run-1/checkpoint-654/special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
run-1/checkpoint-654/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-1/checkpoint-654/tokenizer_config.json
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"100": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"101": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"102": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"103": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": false,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"mask_token": "[MASK]",
|
48 |
+
"model_max_length": 512,
|
49 |
+
"pad_token": "[PAD]",
|
50 |
+
"sep_token": "[SEP]",
|
51 |
+
"strip_accents": null,
|
52 |
+
"tokenize_chinese_chars": true,
|
53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
54 |
+
"unk_token": "[UNK]"
|
55 |
+
}
|
run-1/checkpoint-654/trainer_state.json
ADDED
@@ -0,0 +1,505 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8783473603672532,
|
3 |
+
"best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-1/checkpoint-654",
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 654,
|
7 |
+
"is_hyper_param_search": true,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01529051987767584,
|
13 |
+
"grad_norm": 2.427048683166504,
|
14 |
+
"learning_rate": 7.600520205160823e-05,
|
15 |
+
"loss": 0.5308,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.03058103975535168,
|
20 |
+
"grad_norm": 0.8249368071556091,
|
21 |
+
"learning_rate": 7.482499705080687e-05,
|
22 |
+
"loss": 0.4091,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.045871559633027525,
|
27 |
+
"grad_norm": 3.725421905517578,
|
28 |
+
"learning_rate": 7.36447920500055e-05,
|
29 |
+
"loss": 0.6422,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.06116207951070336,
|
34 |
+
"grad_norm": 2.656184196472168,
|
35 |
+
"learning_rate": 7.246458704920412e-05,
|
36 |
+
"loss": 0.3677,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.0764525993883792,
|
41 |
+
"grad_norm": 3.110931396484375,
|
42 |
+
"learning_rate": 7.128438204840275e-05,
|
43 |
+
"loss": 0.5322,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.09174311926605505,
|
48 |
+
"grad_norm": 6.087479114532471,
|
49 |
+
"learning_rate": 7.010417704760138e-05,
|
50 |
+
"loss": 0.4355,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.10703363914373089,
|
55 |
+
"grad_norm": 0.9337313771247864,
|
56 |
+
"learning_rate": 6.892397204680002e-05,
|
57 |
+
"loss": 0.408,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.12232415902140673,
|
62 |
+
"grad_norm": 6.667540073394775,
|
63 |
+
"learning_rate": 6.774376704599865e-05,
|
64 |
+
"loss": 0.3401,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.13761467889908258,
|
69 |
+
"grad_norm": 3.2563517093658447,
|
70 |
+
"learning_rate": 6.656356204519727e-05,
|
71 |
+
"loss": 0.2753,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1529051987767584,
|
76 |
+
"grad_norm": 0.436352014541626,
|
77 |
+
"learning_rate": 6.53833570443959e-05,
|
78 |
+
"loss": 0.1758,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.16819571865443425,
|
83 |
+
"grad_norm": 2.314504623413086,
|
84 |
+
"learning_rate": 6.420315204359453e-05,
|
85 |
+
"loss": 0.4508,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.1834862385321101,
|
90 |
+
"grad_norm": 5.20156192779541,
|
91 |
+
"learning_rate": 6.302294704279317e-05,
|
92 |
+
"loss": 0.3001,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.19877675840978593,
|
97 |
+
"grad_norm": 0.4600387513637543,
|
98 |
+
"learning_rate": 6.184274204199179e-05,
|
99 |
+
"loss": 0.4469,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.21406727828746178,
|
104 |
+
"grad_norm": 7.823936462402344,
|
105 |
+
"learning_rate": 6.0662537041190425e-05,
|
106 |
+
"loss": 0.535,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.22935779816513763,
|
111 |
+
"grad_norm": 8.010571479797363,
|
112 |
+
"learning_rate": 5.948233204038906e-05,
|
113 |
+
"loss": 0.3132,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.24464831804281345,
|
118 |
+
"grad_norm": 0.605796754360199,
|
119 |
+
"learning_rate": 5.8302127039587676e-05,
|
120 |
+
"loss": 0.4912,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.2599388379204893,
|
125 |
+
"grad_norm": 1.5800836086273193,
|
126 |
+
"learning_rate": 5.712192203878631e-05,
|
127 |
+
"loss": 0.4953,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.27522935779816515,
|
132 |
+
"grad_norm": 5.56485652923584,
|
133 |
+
"learning_rate": 5.594171703798494e-05,
|
134 |
+
"loss": 0.4611,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.290519877675841,
|
139 |
+
"grad_norm": 1.0299450159072876,
|
140 |
+
"learning_rate": 5.476151203718357e-05,
|
141 |
+
"loss": 0.3979,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.3058103975535168,
|
146 |
+
"grad_norm": 3.140592098236084,
|
147 |
+
"learning_rate": 5.35813070363822e-05,
|
148 |
+
"loss": 0.6532,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.3211009174311927,
|
153 |
+
"grad_norm": 7.104299545288086,
|
154 |
+
"learning_rate": 5.240110203558083e-05,
|
155 |
+
"loss": 0.4224,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.3363914373088685,
|
160 |
+
"grad_norm": 0.5627728700637817,
|
161 |
+
"learning_rate": 5.1220897034779465e-05,
|
162 |
+
"loss": 0.2206,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.3516819571865443,
|
167 |
+
"grad_norm": 5.487335681915283,
|
168 |
+
"learning_rate": 5.0040692033978085e-05,
|
169 |
+
"loss": 0.4927,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.3669724770642202,
|
174 |
+
"grad_norm": 2.892604112625122,
|
175 |
+
"learning_rate": 4.886048703317672e-05,
|
176 |
+
"loss": 0.6768,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.382262996941896,
|
181 |
+
"grad_norm": 7.0879807472229,
|
182 |
+
"learning_rate": 4.768028203237535e-05,
|
183 |
+
"loss": 0.3987,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.39755351681957185,
|
188 |
+
"grad_norm": 28.431095123291016,
|
189 |
+
"learning_rate": 4.650007703157398e-05,
|
190 |
+
"loss": 0.3762,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.41284403669724773,
|
195 |
+
"grad_norm": 30.579540252685547,
|
196 |
+
"learning_rate": 4.5319872030772615e-05,
|
197 |
+
"loss": 0.3611,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.42813455657492355,
|
202 |
+
"grad_norm": 0.6805054545402527,
|
203 |
+
"learning_rate": 4.413966702997124e-05,
|
204 |
+
"loss": 0.4075,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.4434250764525994,
|
209 |
+
"grad_norm": 0.8125110864639282,
|
210 |
+
"learning_rate": 4.295946202916987e-05,
|
211 |
+
"loss": 0.3955,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.45871559633027525,
|
216 |
+
"grad_norm": 0.7836199998855591,
|
217 |
+
"learning_rate": 4.17792570283685e-05,
|
218 |
+
"loss": 0.2571,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.4740061162079511,
|
223 |
+
"grad_norm": 1.1527467966079712,
|
224 |
+
"learning_rate": 4.059905202756713e-05,
|
225 |
+
"loss": 0.4111,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.4892966360856269,
|
230 |
+
"grad_norm": 2.705764055252075,
|
231 |
+
"learning_rate": 3.941884702676576e-05,
|
232 |
+
"loss": 0.3557,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.5045871559633027,
|
237 |
+
"grad_norm": 2.6506361961364746,
|
238 |
+
"learning_rate": 3.823864202596439e-05,
|
239 |
+
"loss": 0.39,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.5198776758409785,
|
244 |
+
"grad_norm": 1.7446061372756958,
|
245 |
+
"learning_rate": 3.7058437025163016e-05,
|
246 |
+
"loss": 0.3624,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.5351681957186545,
|
251 |
+
"grad_norm": 2.0617918968200684,
|
252 |
+
"learning_rate": 3.587823202436165e-05,
|
253 |
+
"loss": 0.4086,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.5504587155963303,
|
258 |
+
"grad_norm": 35.361595153808594,
|
259 |
+
"learning_rate": 3.469802702356028e-05,
|
260 |
+
"loss": 0.2898,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.5657492354740061,
|
265 |
+
"grad_norm": 2.274883508682251,
|
266 |
+
"learning_rate": 3.351782202275891e-05,
|
267 |
+
"loss": 0.3872,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.581039755351682,
|
272 |
+
"grad_norm": 4.199743747711182,
|
273 |
+
"learning_rate": 3.233761702195754e-05,
|
274 |
+
"loss": 0.3484,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.5963302752293578,
|
279 |
+
"grad_norm": 4.481986045837402,
|
280 |
+
"learning_rate": 3.115741202115617e-05,
|
281 |
+
"loss": 0.3719,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.6116207951070336,
|
286 |
+
"grad_norm": 1.4046356678009033,
|
287 |
+
"learning_rate": 2.99772070203548e-05,
|
288 |
+
"loss": 0.2779,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.6269113149847095,
|
293 |
+
"grad_norm": 4.377758979797363,
|
294 |
+
"learning_rate": 2.8797002019553428e-05,
|
295 |
+
"loss": 0.3411,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.6422018348623854,
|
300 |
+
"grad_norm": 7.996399402618408,
|
301 |
+
"learning_rate": 2.761679701875206e-05,
|
302 |
+
"loss": 0.4509,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.6574923547400612,
|
307 |
+
"grad_norm": 1.9654350280761719,
|
308 |
+
"learning_rate": 2.6436592017950687e-05,
|
309 |
+
"loss": 0.256,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.672782874617737,
|
314 |
+
"grad_norm": 6.0152435302734375,
|
315 |
+
"learning_rate": 2.525638701714932e-05,
|
316 |
+
"loss": 0.3963,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.6880733944954128,
|
321 |
+
"grad_norm": 1.1808817386627197,
|
322 |
+
"learning_rate": 2.4076182016347952e-05,
|
323 |
+
"loss": 0.2257,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.7033639143730887,
|
328 |
+
"grad_norm": 2.4409139156341553,
|
329 |
+
"learning_rate": 2.2895977015546578e-05,
|
330 |
+
"loss": 0.476,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.7186544342507645,
|
335 |
+
"grad_norm": 2.775444507598877,
|
336 |
+
"learning_rate": 2.171577201474521e-05,
|
337 |
+
"loss": 0.2066,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.7339449541284404,
|
342 |
+
"grad_norm": 1.8437581062316895,
|
343 |
+
"learning_rate": 2.053556701394384e-05,
|
344 |
+
"loss": 0.2967,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.7492354740061162,
|
349 |
+
"grad_norm": 0.8174537420272827,
|
350 |
+
"learning_rate": 1.935536201314247e-05,
|
351 |
+
"loss": 0.2105,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.764525993883792,
|
356 |
+
"grad_norm": 6.47490119934082,
|
357 |
+
"learning_rate": 1.8175157012341098e-05,
|
358 |
+
"loss": 0.2868,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.7798165137614679,
|
363 |
+
"grad_norm": 0.36116883158683777,
|
364 |
+
"learning_rate": 1.699495201153973e-05,
|
365 |
+
"loss": 0.2572,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.7951070336391437,
|
370 |
+
"grad_norm": 3.4261679649353027,
|
371 |
+
"learning_rate": 1.581474701073836e-05,
|
372 |
+
"loss": 0.4464,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.8103975535168195,
|
377 |
+
"grad_norm": 5.803319931030273,
|
378 |
+
"learning_rate": 1.4634542009936987e-05,
|
379 |
+
"loss": 0.3471,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.8256880733944955,
|
384 |
+
"grad_norm": 13.390414237976074,
|
385 |
+
"learning_rate": 1.345433700913562e-05,
|
386 |
+
"loss": 0.305,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.8409785932721713,
|
391 |
+
"grad_norm": 8.899558067321777,
|
392 |
+
"learning_rate": 1.227413200833425e-05,
|
393 |
+
"loss": 0.3923,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.8562691131498471,
|
398 |
+
"grad_norm": 7.016339302062988,
|
399 |
+
"learning_rate": 1.1093927007532877e-05,
|
400 |
+
"loss": 0.2968,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.8715596330275229,
|
405 |
+
"grad_norm": 4.5831427574157715,
|
406 |
+
"learning_rate": 9.91372200673151e-06,
|
407 |
+
"loss": 0.2957,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.8868501529051988,
|
412 |
+
"grad_norm": 1.5068073272705078,
|
413 |
+
"learning_rate": 8.733517005930139e-06,
|
414 |
+
"loss": 0.2239,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.9021406727828746,
|
419 |
+
"grad_norm": 11.313027381896973,
|
420 |
+
"learning_rate": 7.553312005128768e-06,
|
421 |
+
"loss": 0.368,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.9174311926605505,
|
426 |
+
"grad_norm": 6.738920211791992,
|
427 |
+
"learning_rate": 6.373107004327399e-06,
|
428 |
+
"loss": 0.4739,
|
429 |
+
"step": 600
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.9327217125382263,
|
433 |
+
"grad_norm": 4.635502815246582,
|
434 |
+
"learning_rate": 5.192902003526027e-06,
|
435 |
+
"loss": 0.3667,
|
436 |
+
"step": 610
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.9480122324159022,
|
440 |
+
"grad_norm": 2.101200580596924,
|
441 |
+
"learning_rate": 4.012697002724658e-06,
|
442 |
+
"loss": 0.4852,
|
443 |
+
"step": 620
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.963302752293578,
|
447 |
+
"grad_norm": 0.7536319494247437,
|
448 |
+
"learning_rate": 2.8324920019232884e-06,
|
449 |
+
"loss": 0.1926,
|
450 |
+
"step": 630
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.9785932721712538,
|
454 |
+
"grad_norm": 2.7094335556030273,
|
455 |
+
"learning_rate": 1.652287001121918e-06,
|
456 |
+
"loss": 0.3674,
|
457 |
+
"step": 640
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.9938837920489296,
|
461 |
+
"grad_norm": 30.233768463134766,
|
462 |
+
"learning_rate": 4.72082000320548e-07,
|
463 |
+
"loss": 0.3524,
|
464 |
+
"step": 650
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 1.0,
|
468 |
+
"eval_accuracy": 0.8783473603672532,
|
469 |
+
"eval_f1": 0.7527216174183515,
|
470 |
+
"eval_loss": 0.3004890978336334,
|
471 |
+
"eval_precision": 0.7311178247734139,
|
472 |
+
"eval_recall": 0.7756410256410257,
|
473 |
+
"eval_runtime": 4.9159,
|
474 |
+
"eval_samples_per_second": 265.874,
|
475 |
+
"eval_steps_per_second": 16.681,
|
476 |
+
"step": 654
|
477 |
+
}
|
478 |
+
],
|
479 |
+
"logging_steps": 10,
|
480 |
+
"max_steps": 654,
|
481 |
+
"num_input_tokens_seen": 0,
|
482 |
+
"num_train_epochs": 1,
|
483 |
+
"save_steps": 500,
|
484 |
+
"stateful_callbacks": {
|
485 |
+
"TrainerControl": {
|
486 |
+
"args": {
|
487 |
+
"should_epoch_stop": false,
|
488 |
+
"should_evaluate": false,
|
489 |
+
"should_log": false,
|
490 |
+
"should_save": true,
|
491 |
+
"should_training_stop": true
|
492 |
+
},
|
493 |
+
"attributes": {}
|
494 |
+
}
|
495 |
+
},
|
496 |
+
"total_flos": 688830473011200.0,
|
497 |
+
"train_batch_size": 8,
|
498 |
+
"trial_name": null,
|
499 |
+
"trial_params": {
|
500 |
+
"learning_rate": 7.71854070524096e-05,
|
501 |
+
"num_train_epochs": 1,
|
502 |
+
"per_device_train_batch_size": 8,
|
503 |
+
"seed": 4
|
504 |
+
}
|
505 |
+
}
|
run-1/checkpoint-654/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:076f4e50b7e42e2dabdab3aac7da71a30ce1f51982e20a1ad3ab423d6f31bb1b
|
3 |
+
size 5304
|
run-1/checkpoint-654/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-2/checkpoint-492/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
run-2/checkpoint-492/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4288c1f01672e804e837dc6923e20bb8a780538ced5e9f71e36fb446b2ebfc57
|
3 |
+
size 541317368
|
run-2/checkpoint-492/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee80cbb82698e2781ac0155ce7534eb528232b5474a565ae6aa34e588043ce5a
|
3 |
+
size 1082696890
|
run-2/checkpoint-492/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4232552436e4651f6371a0dd4079cf82889ecfe75ebea70442299843daf08cb9
|
3 |
+
size 14308
|
run-2/checkpoint-492/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:048f8be56a6b40d0f9c2d9ee6761b79f64b68cbde8bd277cb478882c33e977e5
|
3 |
+
size 1064
|
run-2/checkpoint-492/special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
run-2/checkpoint-492/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-2/checkpoint-492/tokenizer_config.json
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"100": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"101": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"102": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"103": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": false,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"mask_token": "[MASK]",
|
48 |
+
"model_max_length": 512,
|
49 |
+
"pad_token": "[PAD]",
|
50 |
+
"sep_token": "[SEP]",
|
51 |
+
"strip_accents": null,
|
52 |
+
"tokenize_chinese_chars": true,
|
53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
54 |
+
"unk_token": "[UNK]"
|
55 |
+
}
|
run-2/checkpoint-492/trainer_state.json
ADDED
@@ -0,0 +1,417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.791124713083397,
|
3 |
+
"best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-2/checkpoint-492",
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 492,
|
7 |
+
"is_hyper_param_search": true,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.06097560975609756,
|
13 |
+
"grad_norm": 3.2577075958251953,
|
14 |
+
"learning_rate": 1.7097812500543722e-06,
|
15 |
+
"loss": 0.7201,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.12195121951219512,
|
20 |
+
"grad_norm": 1.7486741542816162,
|
21 |
+
"learning_rate": 1.6743086100117504e-06,
|
22 |
+
"loss": 0.6884,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.18292682926829268,
|
27 |
+
"grad_norm": 2.3384933471679688,
|
28 |
+
"learning_rate": 1.6388359699691285e-06,
|
29 |
+
"loss": 0.6482,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.24390243902439024,
|
34 |
+
"grad_norm": 2.0673463344573975,
|
35 |
+
"learning_rate": 1.6033633299265067e-06,
|
36 |
+
"loss": 0.6254,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.3048780487804878,
|
41 |
+
"grad_norm": 1.9012755155563354,
|
42 |
+
"learning_rate": 1.5678906898838849e-06,
|
43 |
+
"loss": 0.6049,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.36585365853658536,
|
48 |
+
"grad_norm": 1.1691405773162842,
|
49 |
+
"learning_rate": 1.532418049841263e-06,
|
50 |
+
"loss": 0.5601,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.4268292682926829,
|
55 |
+
"grad_norm": 2.259779691696167,
|
56 |
+
"learning_rate": 1.4969454097986412e-06,
|
57 |
+
"loss": 0.5567,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.4878048780487805,
|
62 |
+
"grad_norm": 1.7687610387802124,
|
63 |
+
"learning_rate": 1.4614727697560194e-06,
|
64 |
+
"loss": 0.5499,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.5487804878048781,
|
69 |
+
"grad_norm": 1.2938015460968018,
|
70 |
+
"learning_rate": 1.4260001297133975e-06,
|
71 |
+
"loss": 0.5247,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.6097560975609756,
|
76 |
+
"grad_norm": 1.534727692604065,
|
77 |
+
"learning_rate": 1.3905274896707757e-06,
|
78 |
+
"loss": 0.5265,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.6707317073170732,
|
83 |
+
"grad_norm": 1.6018732786178589,
|
84 |
+
"learning_rate": 1.3550548496281539e-06,
|
85 |
+
"loss": 0.4982,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.7317073170731707,
|
90 |
+
"grad_norm": 1.9590264558792114,
|
91 |
+
"learning_rate": 1.319582209585532e-06,
|
92 |
+
"loss": 0.4593,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.7926829268292683,
|
97 |
+
"grad_norm": 1.6383683681488037,
|
98 |
+
"learning_rate": 1.2841095695429102e-06,
|
99 |
+
"loss": 0.5132,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.8536585365853658,
|
104 |
+
"grad_norm": 1.6192731857299805,
|
105 |
+
"learning_rate": 1.2486369295002884e-06,
|
106 |
+
"loss": 0.4783,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.9146341463414634,
|
111 |
+
"grad_norm": 1.8902620077133179,
|
112 |
+
"learning_rate": 1.2131642894576666e-06,
|
113 |
+
"loss": 0.4747,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.975609756097561,
|
118 |
+
"grad_norm": 2.1500866413116455,
|
119 |
+
"learning_rate": 1.1776916494150447e-06,
|
120 |
+
"loss": 0.5078,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 1.0,
|
125 |
+
"eval_accuracy": 0.7697016067329763,
|
126 |
+
"eval_f1": 0.0960960960960961,
|
127 |
+
"eval_loss": 0.4581395089626312,
|
128 |
+
"eval_precision": 0.7619047619047619,
|
129 |
+
"eval_recall": 0.05128205128205128,
|
130 |
+
"eval_runtime": 4.8871,
|
131 |
+
"eval_samples_per_second": 267.439,
|
132 |
+
"eval_steps_per_second": 16.779,
|
133 |
+
"step": 164
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 1.0365853658536586,
|
137 |
+
"grad_norm": 2.7179384231567383,
|
138 |
+
"learning_rate": 1.1422190093724229e-06,
|
139 |
+
"loss": 0.4381,
|
140 |
+
"step": 170
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.0975609756097562,
|
144 |
+
"grad_norm": 1.8275049924850464,
|
145 |
+
"learning_rate": 1.106746369329801e-06,
|
146 |
+
"loss": 0.446,
|
147 |
+
"step": 180
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.1585365853658536,
|
151 |
+
"grad_norm": 1.6741889715194702,
|
152 |
+
"learning_rate": 1.0712737292871792e-06,
|
153 |
+
"loss": 0.4451,
|
154 |
+
"step": 190
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 1.2195121951219512,
|
158 |
+
"grad_norm": 1.8023988008499146,
|
159 |
+
"learning_rate": 1.0358010892445574e-06,
|
160 |
+
"loss": 0.4538,
|
161 |
+
"step": 200
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.2804878048780488,
|
165 |
+
"grad_norm": 2.3258135318756104,
|
166 |
+
"learning_rate": 1.0003284492019356e-06,
|
167 |
+
"loss": 0.4531,
|
168 |
+
"step": 210
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.3414634146341464,
|
172 |
+
"grad_norm": 2.654583215713501,
|
173 |
+
"learning_rate": 9.648558091593137e-07,
|
174 |
+
"loss": 0.426,
|
175 |
+
"step": 220
|
176 |
+
},
|
177 |
+
{
|
178 |
+
"epoch": 1.4024390243902438,
|
179 |
+
"grad_norm": 2.7538747787475586,
|
180 |
+
"learning_rate": 9.293831691166919e-07,
|
181 |
+
"loss": 0.4056,
|
182 |
+
"step": 230
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.4634146341463414,
|
186 |
+
"grad_norm": 1.887471079826355,
|
187 |
+
"learning_rate": 8.939105290740702e-07,
|
188 |
+
"loss": 0.448,
|
189 |
+
"step": 240
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 1.524390243902439,
|
193 |
+
"grad_norm": 1.7449885606765747,
|
194 |
+
"learning_rate": 8.584378890314483e-07,
|
195 |
+
"loss": 0.4132,
|
196 |
+
"step": 250
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 1.5853658536585367,
|
200 |
+
"grad_norm": 1.4998096227645874,
|
201 |
+
"learning_rate": 8.229652489888265e-07,
|
202 |
+
"loss": 0.4164,
|
203 |
+
"step": 260
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.6463414634146343,
|
207 |
+
"grad_norm": 4.520744800567627,
|
208 |
+
"learning_rate": 7.874926089462047e-07,
|
209 |
+
"loss": 0.4662,
|
210 |
+
"step": 270
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 1.7073170731707317,
|
214 |
+
"grad_norm": 1.9681720733642578,
|
215 |
+
"learning_rate": 7.520199689035828e-07,
|
216 |
+
"loss": 0.3977,
|
217 |
+
"step": 280
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"epoch": 1.7682926829268293,
|
221 |
+
"grad_norm": 3.49233341217041,
|
222 |
+
"learning_rate": 7.16547328860961e-07,
|
223 |
+
"loss": 0.4133,
|
224 |
+
"step": 290
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.8292682926829267,
|
228 |
+
"grad_norm": 2.177455425262451,
|
229 |
+
"learning_rate": 6.810746888183392e-07,
|
230 |
+
"loss": 0.4403,
|
231 |
+
"step": 300
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.8902439024390243,
|
235 |
+
"grad_norm": 3.338192939758301,
|
236 |
+
"learning_rate": 6.456020487757174e-07,
|
237 |
+
"loss": 0.4154,
|
238 |
+
"step": 310
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.951219512195122,
|
242 |
+
"grad_norm": 2.005662441253662,
|
243 |
+
"learning_rate": 6.101294087330955e-07,
|
244 |
+
"loss": 0.4137,
|
245 |
+
"step": 320
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 2.0,
|
249 |
+
"eval_accuracy": 0.7888293802601377,
|
250 |
+
"eval_f1": 0.31,
|
251 |
+
"eval_loss": 0.399868905544281,
|
252 |
+
"eval_precision": 0.7045454545454546,
|
253 |
+
"eval_recall": 0.1987179487179487,
|
254 |
+
"eval_runtime": 4.9013,
|
255 |
+
"eval_samples_per_second": 266.666,
|
256 |
+
"eval_steps_per_second": 16.73,
|
257 |
+
"step": 328
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.0121951219512195,
|
261 |
+
"grad_norm": 1.8187142610549927,
|
262 |
+
"learning_rate": 5.746567686904737e-07,
|
263 |
+
"loss": 0.4231,
|
264 |
+
"step": 330
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 2.073170731707317,
|
268 |
+
"grad_norm": 2.4677720069885254,
|
269 |
+
"learning_rate": 5.391841286478519e-07,
|
270 |
+
"loss": 0.3832,
|
271 |
+
"step": 340
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 2.1341463414634148,
|
275 |
+
"grad_norm": 2.2957379817962646,
|
276 |
+
"learning_rate": 5.0371148860523e-07,
|
277 |
+
"loss": 0.3671,
|
278 |
+
"step": 350
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 2.1951219512195124,
|
282 |
+
"grad_norm": 1.6115466356277466,
|
283 |
+
"learning_rate": 4.682388485626082e-07,
|
284 |
+
"loss": 0.3776,
|
285 |
+
"step": 360
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 2.2560975609756095,
|
289 |
+
"grad_norm": 2.2633934020996094,
|
290 |
+
"learning_rate": 4.327662085199863e-07,
|
291 |
+
"loss": 0.4199,
|
292 |
+
"step": 370
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 2.317073170731707,
|
296 |
+
"grad_norm": 3.485372304916382,
|
297 |
+
"learning_rate": 3.972935684773645e-07,
|
298 |
+
"loss": 0.3784,
|
299 |
+
"step": 380
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.3780487804878048,
|
303 |
+
"grad_norm": 1.9504722356796265,
|
304 |
+
"learning_rate": 3.6182092843474265e-07,
|
305 |
+
"loss": 0.3538,
|
306 |
+
"step": 390
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 2.4390243902439024,
|
310 |
+
"grad_norm": 2.1430771350860596,
|
311 |
+
"learning_rate": 3.263482883921208e-07,
|
312 |
+
"loss": 0.4277,
|
313 |
+
"step": 400
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 2.5,
|
317 |
+
"grad_norm": 1.9325039386749268,
|
318 |
+
"learning_rate": 2.90875648349499e-07,
|
319 |
+
"loss": 0.4083,
|
320 |
+
"step": 410
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 2.5609756097560976,
|
324 |
+
"grad_norm": 1.617710828781128,
|
325 |
+
"learning_rate": 2.5540300830687715e-07,
|
326 |
+
"loss": 0.4297,
|
327 |
+
"step": 420
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 2.6219512195121952,
|
331 |
+
"grad_norm": 3.280532121658325,
|
332 |
+
"learning_rate": 2.1993036826425532e-07,
|
333 |
+
"loss": 0.4056,
|
334 |
+
"step": 430
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 2.682926829268293,
|
338 |
+
"grad_norm": 2.3135297298431396,
|
339 |
+
"learning_rate": 1.8445772822163352e-07,
|
340 |
+
"loss": 0.438,
|
341 |
+
"step": 440
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 2.7439024390243905,
|
345 |
+
"grad_norm": 2.195956230163574,
|
346 |
+
"learning_rate": 1.489850881790117e-07,
|
347 |
+
"loss": 0.4187,
|
348 |
+
"step": 450
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 2.8048780487804876,
|
352 |
+
"grad_norm": 2.371246099472046,
|
353 |
+
"learning_rate": 1.1351244813638987e-07,
|
354 |
+
"loss": 0.3459,
|
355 |
+
"step": 460
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 2.8658536585365852,
|
359 |
+
"grad_norm": 3.243722915649414,
|
360 |
+
"learning_rate": 7.803980809376802e-08,
|
361 |
+
"loss": 0.3754,
|
362 |
+
"step": 470
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 2.926829268292683,
|
366 |
+
"grad_norm": 2.472179412841797,
|
367 |
+
"learning_rate": 4.25671680511462e-08,
|
368 |
+
"loss": 0.4097,
|
369 |
+
"step": 480
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 2.9878048780487805,
|
373 |
+
"grad_norm": 2.2248072624206543,
|
374 |
+
"learning_rate": 7.094528008524367e-09,
|
375 |
+
"loss": 0.3784,
|
376 |
+
"step": 490
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 3.0,
|
380 |
+
"eval_accuracy": 0.791124713083397,
|
381 |
+
"eval_f1": 0.3357664233576642,
|
382 |
+
"eval_loss": 0.386940062046051,
|
383 |
+
"eval_precision": 0.696969696969697,
|
384 |
+
"eval_recall": 0.22115384615384615,
|
385 |
+
"eval_runtime": 4.9814,
|
386 |
+
"eval_samples_per_second": 262.374,
|
387 |
+
"eval_steps_per_second": 16.461,
|
388 |
+
"step": 492
|
389 |
+
}
|
390 |
+
],
|
391 |
+
"logging_steps": 10,
|
392 |
+
"max_steps": 492,
|
393 |
+
"num_input_tokens_seen": 0,
|
394 |
+
"num_train_epochs": 3,
|
395 |
+
"save_steps": 500,
|
396 |
+
"stateful_callbacks": {
|
397 |
+
"TrainerControl": {
|
398 |
+
"args": {
|
399 |
+
"should_epoch_stop": false,
|
400 |
+
"should_evaluate": false,
|
401 |
+
"should_log": false,
|
402 |
+
"should_save": true,
|
403 |
+
"should_training_stop": true
|
404 |
+
},
|
405 |
+
"attributes": {}
|
406 |
+
}
|
407 |
+
},
|
408 |
+
"total_flos": 2071790114979840.0,
|
409 |
+
"train_batch_size": 32,
|
410 |
+
"trial_name": null,
|
411 |
+
"trial_params": {
|
412 |
+
"learning_rate": 1.745253890096994e-06,
|
413 |
+
"num_train_epochs": 3,
|
414 |
+
"per_device_train_batch_size": 32,
|
415 |
+
"seed": 32
|
416 |
+
}
|
417 |
+
}
|
run-2/checkpoint-492/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f480cd110cd7aed1ecf9813a881be3f8605f38e33ceb066267c06832fac0b9e1
|
3 |
+
size 5304
|
run-2/checkpoint-492/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-3/checkpoint-328/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
run-3/checkpoint-328/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b7d72d4db36fdf4eee8a19cc74341ad3d3a9c63ce6bcb32175a46289dfa33e54
|
3 |
+
size 541317368
|
run-3/checkpoint-328/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3ae012ac8ab8cb26d222eaee096dd279cf6e3d78a7e3fe1b58685739642d7343
|
3 |
+
size 1082696890
|
run-3/checkpoint-328/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab8cf7c093dc91f12ea4d3e9b8f2616776606719d84c6266df137b0642fa1b65
|
3 |
+
size 14244
|
run-3/checkpoint-328/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afd2b4a0ff87ce2b1e4da58af48f5b67e7d33011949b0ef45dba2421ea1449dc
|
3 |
+
size 1064
|
run-3/checkpoint-328/special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
run-3/checkpoint-328/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-3/checkpoint-328/tokenizer_config.json
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"100": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"101": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"102": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"103": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": false,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"mask_token": "[MASK]",
|
48 |
+
"model_max_length": 512,
|
49 |
+
"pad_token": "[PAD]",
|
50 |
+
"sep_token": "[SEP]",
|
51 |
+
"strip_accents": null,
|
52 |
+
"tokenize_chinese_chars": true,
|
53 |
+
"tokenizer_class": "DistilBertTokenizer",
|
54 |
+
"unk_token": "[UNK]"
|
55 |
+
}
|
run-3/checkpoint-328/trainer_state.json
ADDED
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8722264728385616,
|
3 |
+
"best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-3/checkpoint-328",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 328,
|
7 |
+
"is_hyper_param_search": true,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.12195121951219512,
|
13 |
+
"grad_norm": 1.4140416383743286,
|
14 |
+
"learning_rate": 9.720486892282597e-06,
|
15 |
+
"loss": 0.5947,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.24390243902439024,
|
20 |
+
"grad_norm": 0.7245877981185913,
|
21 |
+
"learning_rate": 9.477474719975531e-06,
|
22 |
+
"loss": 0.5154,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.36585365853658536,
|
27 |
+
"grad_norm": 0.9261638522148132,
|
28 |
+
"learning_rate": 9.234462547668467e-06,
|
29 |
+
"loss": 0.479,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.4878048780487805,
|
34 |
+
"grad_norm": 1.6241005659103394,
|
35 |
+
"learning_rate": 8.991450375361402e-06,
|
36 |
+
"loss": 0.4377,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.6097560975609756,
|
41 |
+
"grad_norm": 1.7590631246566772,
|
42 |
+
"learning_rate": 8.748438203054338e-06,
|
43 |
+
"loss": 0.4112,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.7317073170731707,
|
48 |
+
"grad_norm": 2.0623185634613037,
|
49 |
+
"learning_rate": 8.505426030747272e-06,
|
50 |
+
"loss": 0.3754,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.8536585365853658,
|
55 |
+
"grad_norm": 1.2114999294281006,
|
56 |
+
"learning_rate": 8.262413858440207e-06,
|
57 |
+
"loss": 0.3731,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.975609756097561,
|
62 |
+
"grad_norm": 1.9598636627197266,
|
63 |
+
"learning_rate": 8.019401686133143e-06,
|
64 |
+
"loss": 0.3791,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 1.0,
|
69 |
+
"eval_accuracy": 0.8607498087222647,
|
70 |
+
"eval_f1": 0.7225609756097561,
|
71 |
+
"eval_loss": 0.33913305401802063,
|
72 |
+
"eval_precision": 0.688953488372093,
|
73 |
+
"eval_recall": 0.7596153846153846,
|
74 |
+
"eval_runtime": 4.8898,
|
75 |
+
"eval_samples_per_second": 267.29,
|
76 |
+
"eval_steps_per_second": 16.77,
|
77 |
+
"step": 82
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 1.0975609756097562,
|
81 |
+
"grad_norm": 1.7537786960601807,
|
82 |
+
"learning_rate": 7.776389513826078e-06,
|
83 |
+
"loss": 0.3125,
|
84 |
+
"step": 90
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 1.2195121951219512,
|
88 |
+
"grad_norm": 4.089900493621826,
|
89 |
+
"learning_rate": 7.533377341519012e-06,
|
90 |
+
"loss": 0.332,
|
91 |
+
"step": 100
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 1.3414634146341464,
|
95 |
+
"grad_norm": 1.488580346107483,
|
96 |
+
"learning_rate": 7.290365169211947e-06,
|
97 |
+
"loss": 0.286,
|
98 |
+
"step": 110
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 1.4634146341463414,
|
102 |
+
"grad_norm": 1.6245107650756836,
|
103 |
+
"learning_rate": 7.0473529969048826e-06,
|
104 |
+
"loss": 0.2952,
|
105 |
+
"step": 120
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 1.5853658536585367,
|
109 |
+
"grad_norm": 2.5228676795959473,
|
110 |
+
"learning_rate": 6.804340824597818e-06,
|
111 |
+
"loss": 0.2956,
|
112 |
+
"step": 130
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 1.7073170731707317,
|
116 |
+
"grad_norm": 5.439332008361816,
|
117 |
+
"learning_rate": 6.561328652290753e-06,
|
118 |
+
"loss": 0.2699,
|
119 |
+
"step": 140
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 1.8292682926829267,
|
123 |
+
"grad_norm": 6.190291404724121,
|
124 |
+
"learning_rate": 6.318316479983688e-06,
|
125 |
+
"loss": 0.2851,
|
126 |
+
"step": 150
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.951219512195122,
|
130 |
+
"grad_norm": 2.851015567779541,
|
131 |
+
"learning_rate": 6.075304307676622e-06,
|
132 |
+
"loss": 0.2729,
|
133 |
+
"step": 160
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 2.0,
|
137 |
+
"eval_accuracy": 0.8684009181331293,
|
138 |
+
"eval_f1": 0.7329192546583851,
|
139 |
+
"eval_loss": 0.29830601811408997,
|
140 |
+
"eval_precision": 0.7108433734939759,
|
141 |
+
"eval_recall": 0.7564102564102564,
|
142 |
+
"eval_runtime": 4.8964,
|
143 |
+
"eval_samples_per_second": 266.931,
|
144 |
+
"eval_steps_per_second": 16.747,
|
145 |
+
"step": 164
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 2.073170731707317,
|
149 |
+
"grad_norm": 1.8465652465820312,
|
150 |
+
"learning_rate": 5.832292135369558e-06,
|
151 |
+
"loss": 0.2559,
|
152 |
+
"step": 170
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 2.1951219512195124,
|
156 |
+
"grad_norm": 1.781294822692871,
|
157 |
+
"learning_rate": 5.5892799630624935e-06,
|
158 |
+
"loss": 0.2523,
|
159 |
+
"step": 180
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 2.317073170731707,
|
163 |
+
"grad_norm": 4.9645256996154785,
|
164 |
+
"learning_rate": 5.346267790755428e-06,
|
165 |
+
"loss": 0.2737,
|
166 |
+
"step": 190
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 2.4390243902439024,
|
170 |
+
"grad_norm": 2.3662397861480713,
|
171 |
+
"learning_rate": 5.103255618448363e-06,
|
172 |
+
"loss": 0.2607,
|
173 |
+
"step": 200
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 2.5609756097560976,
|
177 |
+
"grad_norm": 3.443523645401001,
|
178 |
+
"learning_rate": 4.8602434461412986e-06,
|
179 |
+
"loss": 0.2301,
|
180 |
+
"step": 210
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 2.682926829268293,
|
184 |
+
"grad_norm": 4.988992691040039,
|
185 |
+
"learning_rate": 4.617231273834233e-06,
|
186 |
+
"loss": 0.2451,
|
187 |
+
"step": 220
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"epoch": 2.8048780487804876,
|
191 |
+
"grad_norm": 6.013027667999268,
|
192 |
+
"learning_rate": 4.374219101527169e-06,
|
193 |
+
"loss": 0.2386,
|
194 |
+
"step": 230
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 2.926829268292683,
|
198 |
+
"grad_norm": 4.056751728057861,
|
199 |
+
"learning_rate": 4.131206929220104e-06,
|
200 |
+
"loss": 0.236,
|
201 |
+
"step": 240
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 3.0,
|
205 |
+
"eval_accuracy": 0.8661055853098699,
|
206 |
+
"eval_f1": 0.7422680412371134,
|
207 |
+
"eval_loss": 0.30293241143226624,
|
208 |
+
"eval_precision": 0.6866485013623979,
|
209 |
+
"eval_recall": 0.8076923076923077,
|
210 |
+
"eval_runtime": 4.9202,
|
211 |
+
"eval_samples_per_second": 265.637,
|
212 |
+
"eval_steps_per_second": 16.666,
|
213 |
+
"step": 246
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 3.048780487804878,
|
217 |
+
"grad_norm": 5.498621940612793,
|
218 |
+
"learning_rate": 3.888194756913039e-06,
|
219 |
+
"loss": 0.2761,
|
220 |
+
"step": 250
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 3.1707317073170733,
|
224 |
+
"grad_norm": 2.0462112426757812,
|
225 |
+
"learning_rate": 3.6451825846059735e-06,
|
226 |
+
"loss": 0.2074,
|
227 |
+
"step": 260
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 3.292682926829268,
|
231 |
+
"grad_norm": 4.663333892822266,
|
232 |
+
"learning_rate": 3.402170412298909e-06,
|
233 |
+
"loss": 0.2357,
|
234 |
+
"step": 270
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 3.4146341463414633,
|
238 |
+
"grad_norm": 4.013532638549805,
|
239 |
+
"learning_rate": 3.159158239991844e-06,
|
240 |
+
"loss": 0.2385,
|
241 |
+
"step": 280
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 3.5365853658536586,
|
245 |
+
"grad_norm": 4.440213680267334,
|
246 |
+
"learning_rate": 2.916146067684779e-06,
|
247 |
+
"loss": 0.226,
|
248 |
+
"step": 290
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 3.658536585365854,
|
252 |
+
"grad_norm": 4.662286758422852,
|
253 |
+
"learning_rate": 2.673133895377714e-06,
|
254 |
+
"loss": 0.2129,
|
255 |
+
"step": 300
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 3.7804878048780486,
|
259 |
+
"grad_norm": 4.345000267028809,
|
260 |
+
"learning_rate": 2.4301217230706493e-06,
|
261 |
+
"loss": 0.2366,
|
262 |
+
"step": 310
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 3.902439024390244,
|
266 |
+
"grad_norm": 2.3199422359466553,
|
267 |
+
"learning_rate": 2.1871095507635844e-06,
|
268 |
+
"loss": 0.2462,
|
269 |
+
"step": 320
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 4.0,
|
273 |
+
"eval_accuracy": 0.8722264728385616,
|
274 |
+
"eval_f1": 0.7481146304675717,
|
275 |
+
"eval_loss": 0.3029697835445404,
|
276 |
+
"eval_precision": 0.7065527065527065,
|
277 |
+
"eval_recall": 0.7948717948717948,
|
278 |
+
"eval_runtime": 4.8827,
|
279 |
+
"eval_samples_per_second": 267.68,
|
280 |
+
"eval_steps_per_second": 16.794,
|
281 |
+
"step": 328
|
282 |
+
}
|
283 |
+
],
|
284 |
+
"logging_steps": 10,
|
285 |
+
"max_steps": 410,
|
286 |
+
"num_input_tokens_seen": 0,
|
287 |
+
"num_train_epochs": 5,
|
288 |
+
"save_steps": 500,
|
289 |
+
"stateful_callbacks": {
|
290 |
+
"TrainerControl": {
|
291 |
+
"args": {
|
292 |
+
"should_epoch_stop": false,
|
293 |
+
"should_evaluate": false,
|
294 |
+
"should_log": false,
|
295 |
+
"should_save": true,
|
296 |
+
"should_training_stop": false
|
297 |
+
},
|
298 |
+
"attributes": {}
|
299 |
+
}
|
300 |
+
},
|
301 |
+
"total_flos": 2704984280555520.0,
|
302 |
+
"train_batch_size": 64,
|
303 |
+
"trial_name": null,
|
304 |
+
"trial_params": {
|
305 |
+
"learning_rate": 9.963499064589662e-06,
|
306 |
+
"num_train_epochs": 5,
|
307 |
+
"per_device_train_batch_size": 64,
|
308 |
+
"seed": 9
|
309 |
+
}
|
310 |
+
}
|
run-3/checkpoint-328/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1ba0abad364ab2cb40de43a35618f46f727f31cf408908221b491914804b79c4
|
3 |
+
size 5304
|
run-3/checkpoint-328/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
run-4/checkpoint-1308/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"max_position_embeddings": 512,
|
13 |
+
"model_type": "distilbert",
|
14 |
+
"n_heads": 12,
|
15 |
+
"n_layers": 6,
|
16 |
+
"output_past": true,
|
17 |
+
"pad_token_id": 0,
|
18 |
+
"problem_type": "single_label_classification",
|
19 |
+
"qa_dropout": 0.1,
|
20 |
+
"seq_classif_dropout": 0.2,
|
21 |
+
"sinusoidal_pos_embds": false,
|
22 |
+
"tie_weights_": true,
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.46.2",
|
25 |
+
"vocab_size": 119547
|
26 |
+
}
|
run-4/checkpoint-1308/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ce10587e7c563cda180ab83457698a1bc0b6aecbac909c89e4d1514e9a925734
|
3 |
+
size 541317368
|
run-4/checkpoint-1308/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:443a8c1523e68cdf482c192fd0e444a65b55c1027d1aac986b0a5aefd4724d47
|
3 |
+
size 1082696890
|