You can deploy this model and start earning money today!

dicta-il/
dictalm2.0-AWQ

Name: dictalm2.0-AWQ
Availability: InStock

text-generation-inference

4-bit

awq

Views

No views yet

Adapting LLMs to Hebrew: Unveiling DictaLM 2.0 with Enhanced Vocabulary and Instruction Capabilities

The DictaLM-2.0 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters trained to specialize in Hebrew text.

For full details of this model please read our release blog post or the technical report.

This model contains the AWQ 4-bit quantized version of the base model DictaLM-2.0.

You can view and access the full collection of base/instruct unquantized/quantized versions of DictaLM-2.0 here.

Example Code

Running this code requires less than 5GB of GPU VRAM.

python

from transformers import pipeline

# This loads the model onto the GPU in bfloat16 precision
model = pipeline('text-generation', 'dicta-il/dictalm2.0-AWQ', device_map='cuda')

# Sample few shot examples
prompt = """
עבר: הלכתי
עתיד: אלך

עבר: שמרתי
עתיד: אשמור

עבר: שמעתי
עתיד: אשמע

עבר: הבנתי
עתיד:
"""

print(model(prompt.strip(), do_sample=False, max_new_tokens=4, stop_sequence='\n'))
# [{'generated_text': 'עבר: הלכתי\nעתיד: אלך\n\nעבר: שמרתי\nעתיד: אשמור\n\nעבר: שמעתי\nעתיד: אשמע\n\nעבר: הבנתי\nעתיד: אבין\n\n'}]

Model Architecture

DictaLM-2.0 is based on the Mistral-7B-v0.1 model with the following changes:

An extended tokenizer with 1,000 injected tokens specifically for Hebrew, increasing the compression rate from 5.78 tokens/word to 2.76 tokens/word.
Continued pretraining on over 190B tokens of naturally occuring text, 50% Hebrew and 50% English.

Notice

DictaLM 2.0 is a pretrained base model and therefore does not have any moderation mechanisms.

Citation

If you use this model, please cite:

bibtex

@misc{shmidman2024adaptingllmshebrewunveiling,
      title={Adapting LLMs to Hebrew: Unveiling DictaLM 2.0 with Enhanced Vocabulary and Instruction Capabilities}, 
      author={Shaltiel Shmidman and Avi Shmidman and Amir DN Cohen and Moshe Koppel},
      year={2024},
      eprint={2407.07080},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2407.07080}, 
}