VextLabsinc/gem-ruby

Correct LoRA rank in card: r=64, alpha=128 (was a fleet-wide r=64 assumption)

Correct base line in NOTICE: -> JUWEL Beryl (80L)

Correct base to juwel-beryl (80L); drop false padding claim; fix load class

Fix base_model_name_or_path: dead local path -> juwel-beryl (80L)

model card

weights: adapter_model.safetensors

weights: adapter_config.json

NOTICE

model card

model card

USE_POLICY.md

NOTICE

license

model card

initial commit