File size: 371 Bytes
9028b94 | 1 2 3 4 5 6 7 8 9 | default_stage:
default_modifiers:
QuantizationModifier:
targets: [Linear]
ignore: ['re:.*lm_head', 're:.*embed_tokens', 're:.*layers[.]0[.].*', 're:.*input_layernorm$',
're:.*norm.*', 're:.*shared_experts.*', 're:.*block_sparse_moe[.]gate$', 're:.*router.*',
're:.*post_attention_layernorm$', 're:.*self_attn.*']
scheme: FP8_BLOCK
|