Compressed Model: MilyaShams/Qwen3-1.7B-Pipe_SparseGPT_SmoothQuant_PTQ

This model was compressed using the llmcompressor framework.

Compression Details

• Base Model: Qwen/Qwen3-1.7B
• Experiment Name: Pipe_SparseGPT_SmoothQuant_PTQ
• Recipe / Modifiers Applied:

[SparseGPTModifier(index=None, group=None, start=None, end=None, update=None, initialized_=True, finalized_=True, started_=True, ended_=True, sparsity=0.5, sparsity_profile=None, mask_structure='0:0', owl_m=None, owl_lmbda=None, sequential_update=False, sequential_targets=['Qwen3DecoderLayer'], targets=['Linear'], ignore=[], block_size=128, dampening_frac=0.01, preserve_sparsity_mask=False, offload_hessians=False), SmoothQuantModifier(index=None, group=None, start=None, end=None, update=None, initialized_=True, finalized_=True, started_=True, ended_=True, smoothing_strength=0.5, mappings=[LayerMap(balance_layers=['re:.*q_proj', 're:.*k_proj', 're:.*v_proj'], smooth_layers='re:.*input_layernorm'), LayerMap(balance_layers=['re:.*gate_proj', 're:.*up_proj'], smooth_layers='re:.*post_attention_layernorm')], ignore=[], num_calibration_steps=None, calibration_function=None), QuantizationModifier(config_groups=None, targets=['Linear'], ignore=[], scheme='W8A8', kv_cache_scheme=None, weight_observer=None, input_observer=None, output_observer=None, observer=None, bypass_divisibility_checks=False, index=None, group=None, start=None, end=None, update=None, initialized_=True, finalized_=None, started_=True, ended_=True)]

Note: This model card was automatically generated. All structural modifiers and parameters used during compression are logged above.