Update README.md
Browse files# NPC model
This repo contains the domain-specific base model we've fine-tuned from Mistral-7B using LoRA.
This model can be used to step a NPC in a game or simulation, to make them:
* `say <player1> "Hello adventurer, care to join me in a quest?"`,
* `attack <player2>`
* `greet <player3>`
* and any other actions you pass in the model's prompt (we call them `skills`!).
Our model has been trained on a very specific format of input prompt. To help you interface with it, we've release a [small python client library available on GitHub](https://github.com/gigaxgames/gigax).
🤗 Stay tuned for more models! 🤗
## Prompt format
```txt
- WORLD KNOWLEDGE: A vast open world full of mystery and adventure.
- KNOWN LOCATIONS: Old Town
- NPCS: John the Brave
- CURRENT LOCATION: Old Town: A quiet and peaceful town.
- CURRENT LOCATION ITEMS: Sword
- LAST EVENTS:
Aldren: Say Sword What a fine sword!
- PROTAGONIST NAME: Aldren
- PROTAGONIST PSYCHOLOGICAL PROFILE: Brave and curious
- PROTAGONIST MEMORIES:
Saved the village
Lost a friend
- PROTAGONIST PENDING QUESTS:
Find the ancient artifact
Defeat the evil warlock
- PROTAGONIST ALLOWED ACTIONS:
Attack <character> : Deliver a powerful blow
Aldren:
```
Or, way more conveniently:
```py
llm = AutoModelForCausalLM.from_pretrained(
"TinyLlama/TinyLlama-1.1B-Chat-v1.0", output_attentions=True
)
tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
model = models.Transformers(llm, tokenizer) # type: ignore
# Get the NPC's input
stepper = NPCStepper(model=model)
action = await stepper.get_action(
context=context,
locations=locations,
NPCs=NPCs,
protagonist=protagonist,
items=items,
events=events,
)
```
# Put your input here:
user_input = '''Use this fact to answer the question: Title of each class Trading Symbol(s) Name of each exchange on which registered
Common Stock, Par Value $.01 Per Share MMM New York Stock Exchange
MMM Chicago Stock Exchange, Inc.
1.500% Notes due 2026 MMM26 New York Stock Exchange
1.750% Notes due 2030 MMM30 New York Stock Exchange
1.500% Notes due 2031 MMM31 New York Stock Exchange
Which debt securities are registered to trade on a national securities exchange under 3M's name as of Q2 of 2023?'''
# Simply use your input as the prompt for base models
prompt = user_input
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False).input_ids.to(model.device)
outputs = model.generate(input_ids=inputs, max_length=2048)[0]
answer_start = int(inputs.shape[-1])
pred = tokenizer.decode(outputs[answer_start:], skip_special_tokens=True)
print(f'### User Input:\n{user_input}\n\n### Assistant Output:\n{pred}')
Domain-Specific Tasks
To easily reproduce our results, we have uploaded the filled-in zero/few-shot input instructions and output completions of each domain-specific task: biomedicine-tasks, finance-tasks, and law-tasks.
Note: those filled-in instructions are specifically tailored for models before alignment and do NOT fit for the specific data format required for chat models.
Citation
If you find our work helpful, please cite us:
@inproceedings{
cheng2024adapting,
title={Adapting Large Language Models via Reading Comprehension},
author={Daixuan Cheng and Shaohan Huang and Furu Wei},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=y886UXPEZ0}
}