Downloads · 30 days
274
4% of all-time downloads
RichardErkhov/instruction-pretrain_-_finance-Llama3-8B-gguf
instruction-pretrain_-_finance-Llama3-8B-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
274
4% of all-time downloads
All-time downloads
7.3K
Public
Repo size
92.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf108 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
finance-Llama3-8B - GGUF
| Name | Quant method | Size |
|---|---|---|
| finance-Llama3-8B.Q2_K.gguf | Q2_K | 2.96GB |
| finance-Llama3-8B.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| finance-Llama3-8B.IQ3_S.gguf | IQ3_S | 3.43GB |
| finance-Llama3-8B.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| finance-Llama3-8B.IQ3_M.gguf | IQ3_M | 3.52GB |
| finance-Llama3-8B.Q3_K.gguf | Q3_K | 3.74GB |
| finance-Llama3-8B.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| finance-Llama3-8B.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| finance-Llama3-8B.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| finance-Llama3-8B.Q4_0.gguf | Q4_0 | 4.34GB |
| finance-Llama3-8B.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| finance-Llama3-8B.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| finance-Llama3-8B.Q4_K.gguf | Q4_K | 4.58GB |
| finance-Llama3-8B.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| finance-Llama3-8B.Q4_1.gguf | Q4_1 | 4.78GB |
| finance-Llama3-8B.Q5_0.gguf | Q5_0 | 5.21GB |
| finance-Llama3-8B.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| finance-Llama3-8B.Q5_K.gguf | Q5_K | 5.34GB |
| finance-Llama3-8B.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| finance-Llama3-8B.Q5_1.gguf | Q5_1 | 5.65GB |
| finance-Llama3-8B.Q6_K.gguf | Q6_K | 6.14GB |
| finance-Llama3-8B.Q8_0.gguf | Q8_0 | 7.95GB |
license: llama3 language:
This repo contains the finance model developed from Llama3-8B in our paper Instruction Pre-Training: Language Models are Supervised Multitask Learners.
We explore supervised multitask pre-training by proposing Instruction Pre-Training, a framework that scalably augments massive raw corpora with instruction-response pairs to pre-train language models. The instruction-response pairs are generated by an efficient instruction synthesizer built on open-source models. Instruction Pre-Training outperforms Vanilla Pre-training in both general pre-training from scratch and domain-adaptive continual pre-training. In pre-training from scratch, Instruction Pre-Training not only improves pre-trained base models but also benefits more from further instruction tuning. In continual pre-training, Instruction Pre-Training enables Llama3-8B to be comparable to or even outperform Llama3-70B.
<p align='center'> <img src="https://cdn-uploads.huggingface.co/production/uploads/66711d2ee12fa6cc5f5dfc89/vRdsFIVQptbNaGiZ18Lih.png" width="400"> </p>**************************** Updates ****************************
Advanced Usage section of instruction-synthesizer🤗 We share our data and models with example usages, feel free to open any discussions at this page! 🤗
Following AdaptLLM, we augment the domain-specific raw corpora with instruction-response pairs generated by our context-based instruction synthesizer.
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("instruction-pretrain/finance-Llama3-8B")
tokenizer = AutoTokenizer.from_pretrained("instruction-pretrain/finance-Llama3-8B")
# Put your input here, NO prompt template is required
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?'''
inputs = tokenizer(user_input, return_tensors="pt", add_special_tokens=True).input_ids.to(model.device)
outputs = model.generate(input_ids=inputs, max_new_tokens=400)[0]
answer_start = int(inputs.shape[-1])
pred = tokenizer.decode(outputs[answer_start:], skip_special_tokens=True)
print(pred)
git clone https://github.com/microsoft/LMOps
cd LMOps/adaptllm
pip install -r requirements.txt
DOMAIN='finance'
# if the model can fit on a single GPU: set MODEL_PARALLEL=False
# elif the model is too large to fit on a single GPU: set MODEL_PARALLEL=True
MODEL_PARALLEL=False
# number of GPUs, chosen from [1,2,4,8]
N_GPU=1
# Set as True
add_bos_token=True
bash scripts/inference.sh ${DOMAIN} 'instruction-pretrain/finance-Llama3-8B' ${add_bos_token} ${MODEL_PARALLEL} ${N_GPU}
If you find our work helpful, please cite us:
Instruction Pre-Training
@article{cheng2024instruction,
title={Instruction Pre-Training: Language Models are Supervised Multitask Learners},
author={Cheng, Daixuan and Gu, Yuxian and Huang, Shaohan and Bi, Junyu and Huang, Minlie and Wei, Furu},
journal={arXiv preprint arXiv:2406.14491},
year={2024}
}
@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}
}