mikhail-panzo
commited on
Commit
•
2831b6f
1
Parent(s):
798fc01
End of training
Browse files- README.md +76 -0
- generation_config.json +9 -0
README.md
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
base_model: mikhail-panzo/zlm_b64_le4_s12000
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
model-index:
|
7 |
+
- name: zlm-fil_b64_le5_s8000
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# zlm-fil_b64_le5_s8000
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [mikhail-panzo/zlm_b64_le4_s12000](https://huggingface.co/mikhail-panzo/zlm_b64_le4_s12000) on an unknown dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.4118
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
More information needed
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Training and evaluation data
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training procedure
|
33 |
+
|
34 |
+
### Training hyperparameters
|
35 |
+
|
36 |
+
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 1e-05
|
38 |
+
- train_batch_size: 32
|
39 |
+
- eval_batch_size: 8
|
40 |
+
- seed: 42
|
41 |
+
- gradient_accumulation_steps: 2
|
42 |
+
- total_train_batch_size: 64
|
43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
+
- lr_scheduler_type: linear
|
45 |
+
- lr_scheduler_warmup_steps: 2000
|
46 |
+
- training_steps: 8000
|
47 |
+
- mixed_precision_training: Native AMP
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
+
|:-------------:|:--------:|:----:|:---------------:|
|
53 |
+
| 0.5529 | 22.2222 | 500 | 0.5000 |
|
54 |
+
| 0.4974 | 44.4444 | 1000 | 0.4557 |
|
55 |
+
| 0.4716 | 66.6667 | 1500 | 0.4359 |
|
56 |
+
| 0.453 | 88.8889 | 2000 | 0.4246 |
|
57 |
+
| 0.4428 | 111.1111 | 2500 | 0.4196 |
|
58 |
+
| 0.4332 | 133.3333 | 3000 | 0.4171 |
|
59 |
+
| 0.4246 | 155.5556 | 3500 | 0.4154 |
|
60 |
+
| 0.4202 | 177.7778 | 4000 | 0.4133 |
|
61 |
+
| 0.4223 | 200.0 | 4500 | 0.4145 |
|
62 |
+
| 0.4127 | 222.2222 | 5000 | 0.4118 |
|
63 |
+
| 0.418 | 244.4444 | 5500 | 0.4130 |
|
64 |
+
| 0.4137 | 266.6667 | 6000 | 0.4130 |
|
65 |
+
| 0.4105 | 288.8889 | 6500 | 0.4127 |
|
66 |
+
| 0.4164 | 311.1111 | 7000 | 0.4127 |
|
67 |
+
| 0.4088 | 333.3333 | 7500 | 0.4120 |
|
68 |
+
| 0.4028 | 355.5556 | 8000 | 0.4118 |
|
69 |
+
|
70 |
+
|
71 |
+
### Framework versions
|
72 |
+
|
73 |
+
- Transformers 4.41.0.dev0
|
74 |
+
- Pytorch 2.2.1+cu121
|
75 |
+
- Datasets 2.19.0
|
76 |
+
- Tokenizers 0.19.1
|
generation_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"decoder_start_token_id": 2,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"max_length": 1876,
|
7 |
+
"pad_token_id": 1,
|
8 |
+
"transformers_version": "4.41.0.dev0"
|
9 |
+
}
|