Instructions to use P1sc3s007/llm4decompile-pt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use P1sc3s007/llm4decompile-pt with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("LLM4Binary/llm4decompile-1.3b-v1.5") model = PeftModel.from_pretrained(base_model, "P1sc3s007/llm4decompile-pt") - Notebooks
- Google Colab
- Kaggle
Download tokenizer.json from P1sc3s007/llm4decompile-pt: direct link, hf CLI and curl.
- Browser
- Download file 1.37 MB
-
https://huggingface.co/P1sc3s007/llm4decompile-pt/resolve/main/tokenizer.json
- Command line
-
hf download hf://P1sc3s007/llm4decompile-pt/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/P1sc3s007/llm4decompile-pt/resolve/main/tokenizer.json
1.37 MB
File too large to display, you can check the raw version instead.