Qwopus3.6-27B-v2-RYS-EQ

This is a RYS/relayer export of Jackrong/Qwopus3.6-27B-v2.

The checkpoint physically duplicates selected decoder layers in the Hugging Face safetensors weights. It does not require the RYS runtime wrapper.

Variant

• Target repo: hampsonw/Qwopus3.6-27B-v2-RYS-EQ
• Objective: EQ-focused
• Repeated block: 14,30 (--blocks "14,30")
• Repeated source layers: 14–29 inclusive, zero-indexed
• Source text layers: 64
• Target text layers: 80
• Extra repeated layers: 16

Probe result

From the BF16 Transformers scan over math_16 + eq_16:

Metric	Score	Delta vs baseline
Math	0.736822	+0.023567
EQ	0.752974	+0.021474

Rank note: EQ rank 1; balanced rank 6.

Baseline scores from the scan:

• Math: 0.713255
• EQ: 0.731500

Provenance

• Source model: Jackrong/Qwopus3.6-27B-v2
• Export method: RYS physical layer duplication
• Export manifest: rys_export_manifest.json
• Probe result bundle: qwopus36-bf16-results_20260523T213305Z.tar.zst

Notes

This model has not yet been validated on the larger math_120 + eq_140 probe set. Treat these as probe-selected candidates pending larger validation.