Upload folder using huggingface_hub

731a7da verified 3 months ago

223 Bytes

quant_stage:
  quant_modifiers:
    GPTQModifier:
      ignore: [lm_head]
      config_groups:
        group_0:
          weights: {num_bits: 4, strategy: group, group_size: 128, dynamic: false}
          targets: [Linear]

	quant_stage:
	quant_modifiers:
	GPTQModifier:
	ignore: [lm_head]
	config_groups:
	group_0:
	weights: {num_bits: 4, strategy: group, group_size: 128, dynamic: false}
	targets: [Linear]