To reproduce this run:

accelerate launch --multi_gpu --mixed_precision=fp16 --num_processes=8 run_distillation.py config_mistral.yaml
Downloads last month
56
Safetensors
Model size
1.57B params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Dataset used to train sanchit-gandhi/distil-mistral-1.5B-v0.1