|
--- |
|
base_model: llm-jp/llm-jp-1.3b-v1.0 |
|
license: apache-2.0 |
|
model_creator: Toshihiko Aoki |
|
model_name: llm-jp-1.3b-v1.0 staircaptions SFT - GGUF |
|
model_type: llm-jp |
|
prompt_template: '### Instruction: {prompt} ### Response: ' |
|
datasets: |
|
- taoki/stair-captions-prompts |
|
--- |
|
|
|
# llm-jp-1.3b-v1.0 staircaptions SFT - GGUF |
|
|
|
## Description |
|
|
|
This repository contains models trained(QLoRA-SFT) based on the following models and data: |
|
- Base model: [llm-jp-1.3b-v.10](https://huggingface.co/llm-jp/llm-jp-1.3b-v1.0) |
|
- Demo: [Spaces(Q2_K model)](https://huggingface.co/spaces/taoki/tiny-ja-trans-sd) |
|
- Training data (split train use): |
|
- [STAIR Captions](https://github.com/STAIR-Lab-CIT/STAIR-captions) + [Microsoft COCO Caption](https://github.com/tylin/coco-caption) + [Microsoft COCO Images caption interpreted by UForm](https://huggingface.co/unum-cloud/uform-gen) |
