Stellar Bright is a general capability upgrade to Llama 2, using open source data to improve overall knowledge, extended communication, and technical skill.

This model is primarily recommended as a superior-to-Llama-2 baseline for additional finetuning, not for direct deployment to production as a chat model. The user accepts full responsibility for all outputs.

This is a 'legacy model' offered primarily for reference purposes. I recommend Llama 3 over this model for general use.

Evaluation

Model Avg ARC HS MMLU TQA
Stellar Bright 74.10 72.95 87.82 71.17 64.46
Llama 2 67.35 67.32 87.33 69.83 44.92
Llama 2 Chat 66.80 64.59 85.88 63.91 52.80
Downloads last month
332
Safetensors
Model size
69B params
Tensor type
F32
Β·

Spaces using sequelbox/Llama2-70B-StellarBright 3