RichardErkhov/AstroMLab_-_astrollama-3-8b-base_aic-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

astrollama-3-8b-base_aic - GGUF

• Model creator: https://huggingface.co/AstroMLab/
• Original model: https://huggingface.co/AstroMLab/astrollama-3-8b-base_aic/

Name	Quant method	Size
astrollama-3-8b-base_aic.Q2_K.gguf	Q2_K	2.96GB
astrollama-3-8b-base_aic.Q3_K_S.gguf	Q3_K_S	3.41GB
astrollama-3-8b-base_aic.Q3_K.gguf	Q3_K	3.74GB
astrollama-3-8b-base_aic.Q3_K_M.gguf	Q3_K_M	3.74GB
astrollama-3-8b-base_aic.Q3_K_L.gguf	Q3_K_L	2.44GB
astrollama-3-8b-base_aic.IQ4_XS.gguf	IQ4_XS	3.28GB
astrollama-3-8b-base_aic.Q4_0.gguf	Q4_0	4.34GB
astrollama-3-8b-base_aic.IQ4_NL.gguf	IQ4_NL	4.38GB
astrollama-3-8b-base_aic.Q4_K_S.gguf	Q4_K_S	4.37GB
astrollama-3-8b-base_aic.Q4_K.gguf	Q4_K	4.58GB
astrollama-3-8b-base_aic.Q4_K_M.gguf	Q4_K_M	4.58GB
astrollama-3-8b-base_aic.Q4_1.gguf	Q4_1	4.78GB
astrollama-3-8b-base_aic.Q5_0.gguf	Q5_0	5.21GB
astrollama-3-8b-base_aic.Q5_K_S.gguf	Q5_K_S	5.21GB
astrollama-3-8b-base_aic.Q5_K.gguf	Q5_K	5.34GB
astrollama-3-8b-base_aic.Q5_K_M.gguf	Q5_K_M	5.34GB
astrollama-3-8b-base_aic.Q5_1.gguf	Q5_1	5.65GB
astrollama-3-8b-base_aic.Q6_K.gguf	Q6_K	6.14GB
astrollama-3-8b-base_aic.Q8_0.gguf	Q8_0	7.95GB

Original model description:

license: mit language: - en pipeline_tag: text-generation tags: - llama-3 - astronomy - astrophysics - arxiv inference: false base_model: - meta-llama/Llama-3-8b-hf

AstroLLaMA-3-8B-Base_AIC

AstroLLaMA-3-8B is a specialized base language model for astronomy, developed by fine-tuning Meta's LLaMA-3-8b architecture on astronomical literature. This model was developed by the AstroMLab team. It is designed for next token prediction tasks and is not an instruct/chat model.

Model Details

• Base Architecture: LLaMA-3-8b
• Training Data: Abstract, Introduction, and Conclusion (AIC) sections from arXiv's astro-ph category papers
• Data Processing: Optical character recognition (OCR) on PDF files using the Nougat tool, followed by summarization using Qwen-2-8B and LLaMA-3.1-8B.
• Fine-tuning Method: Continual Pre-Training (CPT) using the LMFlow framework
• Training Details:
  • Learning rate: 2 × 10⁻⁵
  • Total batch size: 96
  • Maximum token length: 512
  • Warmup ratio: 0.03
  • No gradient accumulation
  • BF16 format
  • Cosine decay schedule for learning rate reduction
  • Training duration: 1 epoch
• Primary Use: Next token prediction for astronomy-related text generation and analysis
• Reference: Pan et al. 2024 [Link to be added]

Generating text from a prompt

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load the model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("AstroMLab/astrollama-3-8b-base_aic")
model = AutoModelForCausalLM.from_pretrained("AstroMLab/astrollama-3-8b-base_aic", device_map="auto")

# Create the pipeline with explicit truncation
from transformers import pipeline
generator = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    device_map="auto",
    truncation=True,
    max_length=512
)

# Example prompt from an astronomy paper
prompt = "In this letter, we report the discovery of the highest redshift, " \
    "heavily obscured, radio-loud QSO candidate selected using JWST NIRCam/MIRI, " \
    "mid-IR, sub-mm, and radio imaging in the COSMOS-Web field. "

# Set seed for reproducibility
torch.manual_seed(42)

# Generate text
generated_text = generator(prompt, do_sample=True)
print(generated_text[0]['generated_text'])

Model Limitations and Biases

A key limitation identified during the development of this model is that training solely on astro-ph data may not be sufficient to significantly improve performance over the base model, especially for the already highly performant LLaMA-3 series. This suggests that to achieve substantial gains, future iterations may need to incorporate a broader range of high-quality astronomical data beyond arXiv, such as textbooks, Wikipedia, and curated summaries.

Here's a performance comparison chart based upon the astronomical benchmarking Q&A as described in Ting et al. 2024, and Pan et al. 2024:

Model	Score (%)
LLaMA-3.1-8B	73.7
LLaMA-3-8B	72.9
AstroLLaMA-3-8B-Base_AIC (AstroMLab)	71.9
Gemma-2-9B	71.5
Qwen-2.5-7B	70.4
Yi-1.5-9B	68.4
InternLM-2.5-7B	64.5
Mistral-7B-v0.3	63.9
ChatGLM3-6B	50.4
AstroLLaMA-2-7B-AIC	44.3
AstroLLaMA-2-7B-Abstract	43.5

As shown, while AstroLLaMA-3-8B performs competitively among models in its class, it does not surpass the performance of the base LLaMA-3-8B model. This underscores the challenges in developing specialized models and the need for more diverse and comprehensive training data.

This model is released primarily for reproducibility purposes, allowing researchers to track the development process and compare different iterations of AstroLLaMA models.

For optimal performance and the most up-to-date capabilities in astronomy-related tasks, we recommend using AstroSage-8B, where these limitations have been addressed. The newer model incorporates expanded training data beyond astro-ph and features a greatly expanded fine-tuning process, resulting in significantly improved performance.

Ethical Considerations

While this model is designed for scientific use, users should be mindful of potential misuse, such as generating misleading scientific content. Always verify model outputs against peer-reviewed sources for critical applications.

Citation

If you use this model in your research, please cite:

[Citation for Pan et al. 2024 to be added]