File size: 1,249 Bytes
735c370
87e97e9
1d24557
 
87e97e9
 
1d24557
 
 
87e97e9
 
 
 
 
 
735c370
1d24557
 
 
 
87e97e9
1d24557
87e97e9
1d24557
87e97e9
1d24557
 
 
 
 
 
 
 
 
 
 
 
 
 
 
87e97e9
1d24557
 
 
 
 
 
87e97e9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
---
license: apache-2.0
tags:
- generated_from_trainer
- KoRWKV
- KoAlpaca
model-index:
- name: KoRWKV-6B-koalpaca-v1.1a
  results: []
datasets:
- beomi/KoAlpaca-v1.1a
language:
- ko
library_name: transformers
pipeline_tag: text-generation
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# KoAlpaca-KoRWKV-6B (v1.1a)

This model is a fine-tuned version of [beomi/KoRWKV-6B](https://huggingface.co/beomi/KoRWKV-6B) on an [KoAlpaca v1.1a Dataset](https://huggingface.co/datasets/beomi/KoAlpaca-v1.1a).

Detail Codes are available at [KoAlpaca Github Repository](https://github.com/Beomi/KoAlpaca)

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1.0
- mixed_precision_training: Native AMP
- Trained on 1x H100(80G PCI-E) GPU

### Framework versions

- Transformers 4.29.2
- Pytorch 1.13.1
- Datasets 2.12.0
- Tokenizers 0.13.3