A series of small models (0.5B-4B) distilled from their 8B model using logit distillation.
Specs
Params4B
LicenseApache-2.0
Tags
modelsmodels/llmsmultilingualartifacts/22
Related Models
A series of small models (0.5B-4B) distilled from their 8B model using logit distillation.