Files
Meta-Llama-3-8B-Instruct-qu…/recipe.yaml

12 lines
381 B
YAML
Raw Normal View History

quant_stage:
quant_modifiers:
GPTQModifier:
sequential_update: false
dampening_frac: 0.01
ignore: [lm_head]
config_groups:
group_0:
targets: [Linear]
weights: {num_bits: 8, type: int, symmetric: true, strategy: channel}
input_activations: {num_bits: 8, type: int, symmetric: true, dynamic: true, strategy: token}