Instructions to use CMU-AIR2/math-deepseek-lora-arith-all-simple with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use CMU-AIR2/math-deepseek-lora-arith-all-simple with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-coder-1.3b-instruct") model = PeftModel.from_pretrained(base_model, "CMU-AIR2/math-deepseek-lora-arith-all-simple") - Notebooks
- Google Colab
- Kaggle
Commit History
Add tokenizer 8938965
Fred commited on
Add config.json from https://huggingface.co/CMU-AIR2/deepseek_lora_hardarithmetic/commit/897694fea18d122a157c31409e5455b3adc7ccd6 d1fd2f5
Fred commited on
Can only hold one checkpoint?? 5ab41a0
Fred commited on
Save config for this run a7bb718
Fred commited on
180K steps (WandB run: https://wandb.ai/cmu-11785-s24-sg06/huggingface/runs/3060wpp0 at 80K steps) d5dd255
Fred commited on