Bonito-v1-awq Open-source Model - Convert Unlabeled Texts into Training Datasets for Specific Tasks! Free to Use

Bonito V1 Awq

Developed by alexandreteles

Bonito is an open-source 'conditional task generation' model capable of transforming unannotated text into task-specific instruction-tuning training datasets.

Large Language Model

Transformers

EnglishOpen Source License:Apache-2.0 #Task Dataset Generation #Unsupervised Instruction Tuning #Mistral Architecture Optimization

Downloads 203

Release Time : 2/27/2024

Model Overview

Bonito is used to create synthetic instruction datasets, helping large language models adapt to users' specialized private data. The model enables pre-trained and instruction-tuned models to adapt to new tasks without requiring any annotations.

Model Features

Conditional Task Generation

Capable of transforming unannotated text into task-specific instruction-tuning training datasets

No Annotation Required

Enables model adaptation to new tasks without requiring any annotations

Multi-task Support

Supports generating instruction datasets for multiple task types

Model Capabilities

Text Generation

Instruction Dataset Generation

Task Adaptation

Use Cases

Data Augmentation

Generate Synthetic Instruction Datasets

Generate training data for specific tasks

Helps large language models adapt to specialized private data

Model Adaptation

Task Adaptation

Enable pre-trained models to adapt to new tasks

Achieves task adaptation without requiring annotations

🚀 Bonito-v1 AWQ

Bonito-v1 AWQ is a model for conditional task generation, which can convert unannotated text into task - specific training datasets for instruction tuning.

🚀 Quick Start

You can find the original model at BatsResearch/bonito-v1

✨ Features

Variations

GEMM: model.safetensors
GEMV: model_gemv.safetensors

📚 Documentation

Model Card for bonito

Bonito is an open - source model for conditional task generation: the task of converting unannotated text into task - specific training datasets for instruction tuning.

Bonito

Model Details

Model Description

Bonito can be used to create synthetic instruction tuning datasets to adapt large language models on users' specialized, private data. In our paper, we show that Bonito can be used to adapt both pretrained and instruction tuned models to tasks without any annotations.

Property	Details
Developed by	Nihal V. Nayak, Yiyang Nan, Avi Trost, and Stephen H. Bach
Model Type	MistralForCausalLM
Language(s) (NLP)	English
License	TBD
Finetuned from model	`mistralai/Mistral-7B-v0.1`

Model Sources

Repository: https://github.com/BatsResearch/bonito
Paper: Arxiv link

Uses

Direct Use

To easily generate synthetic instruction tuning datasets, we recommend using the bonito package built using the transformers and the vllm libraries.

from bonito import Bonito, SamplingParams
from datasets import load_dataset

# Initialize the Bonito model
bonito = Bonito()

# load dataaset with unannotated text
unannotated_text = load_dataset(
    "BatsResearch/bonito-experiment",
    "unannotated_contract_nli"
)["train"].select(range(10))

# Generate synthetic instruction tuning dataset
sampling_params = SamplingParams(max_tokens=256, top_p=0.95, temperature=0.5, n=1)
synthetic_dataset = bonito.generate_tasks(
    unannotated_text,
    context_col="input",
    task_type="nli",
    sampling_params=sampling_params
)

Out - of - Scope Use

Our model is trained to generate the following task types: summarization, sentiment analysis, multiple - choice question answering, extractive question answering, topic classification, natural language inference, question generation, text generation, question answering without choices, paraphrase identification, sentence completion, yes - no question answering, word sense disambiguation, paraphrase generation, textual entailment, and coreference resolution. The model might not produce accurate synthetic tasks beyond these task types.

📄 License

apache - 2.0

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご