Views
No views yet
outlier (4-bit-class)Qwen/Qwen2.5-32B-Instruct, produced by the project's Tail-Aware
Quantization (TAQ) pipeline — uniform outlier action applied to every eligible layer, at a ~4-bit budget.| bpw | PPL | KLD vs fp16 | |
|---|---|---|---|
This checkpoint (outlier_4bit, uniform) | 4.7594 | 6.060 | 0.1683 |
rotation_outlier_4bit (uniform, same budget class) | 4.7784 | 6.377 | 0.2170 |
mixed_allocator_4bit (per-layer allocator, same budget class) | 4.7581 | 5.996 | 0.1619 |
outlier beat rotation_outlier at this budget (0.1683 vs 0.2170 KLD).rshoemake/taq-qwen14b-unsloth-matched:
packed/*.bin per-layer files (header + fp16 codebooks + bit-packed
indices + outlier side-channel), packed/manifest.json (per-layer
action/bits/measured bpw), fp16_passthrough/*.safetensors for any
layers the allocator left at full precision, unpack_outlier.py /
unpack_rotation_outlier.py standalone unpackers.Qwen/Qwen2.5-32B-Instruct (fp16), packed directly from the
original safetensors shards.run_full_matrix.py / run_tail_family_scale.py
(project fractal_quant_decision, private) — native family-scale
protocol (WikiText-2, passage-level KLD).pack_outlier_format.py + pack_rotation_outlier.py +
pack_whole_model.py.