|
--- |
|
license: apache-2.0 |
|
language: |
|
- en |
|
- ja |
|
tags: |
|
- finetuned |
|
library_name: transformers |
|
pipeline_tag: text-generation |
|
--- |
|
<img src="./ninjalogo.svg" width="100%" height="20%" alt=""> |
|
|
|
# Our Models |
|
- [Vecteus](https://huggingface.co/Local-Novel-LLM-project/Vecteus-v1) |
|
|
|
- [Ninja-v1](https://huggingface.co/Local-Novel-LLM-project/Ninja-v1) |
|
|
|
- [Ninja-v1-NSFW](https://huggingface.co/Local-Novel-LLM-project/Ninja-v1-NSFW) |
|
|
|
- [Ninja-v1-128k](https://huggingface.co/Local-Novel-LLM-project/Ninja-v1-128k) |
|
|
|
- [Ninja-v1-NSFW-128k](https://huggingface.co/Local-Novel-LLM-project/Ninja-v1-NSFW-128k) |
|
|
|
## Model Card for Ninja-v1.0 |
|
|
|
The Mistral-7B--based Large Language Model (LLM) is an noveldataset fine-tuned version of the Mistral-7B-v0.1 |
|
|
|
Ninja has the following changes compared to Mistral-7B-v0.1. |
|
- Achieving both high quality Japanese and English generation |
|
- Memory ability that does not forget even after long-context generation |
|
|
|
This model was created with the help of GPUs from the first LocalAI hackathon. |
|
|
|
We would like to take this opportunity to thank |
|
|
|
## List of Creation Methods |
|
|
|
- Chatvector for multiple models |
|
- Simple linear merging of result models |
|
- Domain and Sentence Enhancement with LORA |
|
- Context expansion |
|
|
|
## Instruction format |
|
|
|
Ninja adopts the prompt format from Vicuna and supports multi-turn conversation. |
|
The prompt should be as following: |
|
``` |
|
USER: Hi ASSISTANT: Hello.</s> |
|
USER: Who are you? |
|
ASSISTANT: I am ninja.</s> |
|
``` |
|
|
|
## Example prompts to improve (Japanese) |
|
|
|
- BAD:ใใใชใใฏโโใจใใฆๆฏใ่ใใพใ |
|
- GOOD: ใใชใใฏโโใงใ |
|
|
|
- BAD: ใใชใใฏโโใใงใใพใ |
|
- GOOD: ใใชใใฏโโใใใพใ |
|
|
|
## Performing inference |
|
|
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
import torch |
|
|
|
model_id = "Local-Novel-LLM-project/Ninja-v1" |
|
new_tokens = 1024 |
|
|
|
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, torch_dtype=torch.float16, attn_implementation="flash_attention_2", device_map="auto") |
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
|
|
system_prompt = "ใใชใใฏใใญใฎๅฐ่ชฌๅฎถใงใใ\nๅฐ่ชฌใๆธใใฆใใ ใใ\n-------- " |
|
|
|
prompt = input("Enter a prompt: ") |
|
system_prompt += prompt + "\n-------- " |
|
model_inputs = tokenizer([system_prompt], return_tensors="pt").to("cuda") |
|
|
|
|
|
generated_ids = model.generate(**model_inputs, max_new_tokens=new_tokens, do_sample=True) |
|
print(tokenizer.batch_decode(generated_ids)[0]) |
|
```` |
|
|
|
## Merge recipe |
|
|
|
- WizardLM2 - mistralai/Mistral-7B-v0.1 |
|
- Elizezen/Antler-7B - stabilityai/japanese-stablelm-instruct-gamma-7b |
|
- NTQAI/chatntq-ja-7b-v1.0 |
|
|
|
The characteristics of each model are as follows. |
|
|
|
- WizardLM2: High quality multitasking model |
|
- Antler-7B: Model specialized for novel writing |
|
- NTQAI/chatntq-ja-7b-v1.0 High quality Japanese specialized model |
|
|
|
## Other points to keep in mind |
|
- The training data may be biased. Be careful with the generated sentences. |
|
- Memory usage may be large for long inferences. |
|
- If possible, we recommend inferring with llamacpp rather than Transformers. |
