ThinkMix 1B

Fine-tuned on the use of <think></think> tags for reasoning.

Works best around temperature 0.5-0.6.

  • Developed by: theprint
  • License: apache-2.0
  • Finetuned from model : unsloth/llama-3.2-1b-instruct-unsloth-bnb-4bit

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month
511
Safetensors
Model size
1B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for theprint/Llama3.2-1B-ThinkMix

Quantizations
3 models

Dataset used to train theprint/Llama3.2-1B-ThinkMix