RichardErkhov/LSX-UniWue_-_LLaMmlein_120M-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

LLaMmlein_120M - bnb 8bits

• Model creator: https://huggingface.co/LSX-UniWue/
• Original model: https://huggingface.co/LSX-UniWue/LLaMmlein_120M/

Original model description:

datasets: - togethercomputer/RedPajama-Data-V2 language: - de pipeline_tag: text-generation library_name: transformers license: other

LLäMmlein 120M

This is a German Tinyllama 120M language model trained from scratch using the Tinyllama codebase on the German portion of RedPajama V2. Find more details on our page and our preprint!

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("LSX-UniWue/LLaMmlein_120M")

tokenizer = AutoTokenizer.from_pretrained("LSX-UniWue/LLaMmlein_120M")

Performance

We evaluated our model on the SuperGLEBer benchmark.