Downloads · 30 days
14
33% of all-time downloads
albedo-qwen3.6-35b-bookend-phase4-sft
JMaxCool/albedo-qwen3.6-35b-bookend-phase4-sft
albedo-qwen3.6-35b-bookend-phase4-sft is a machine learning model from JMaxCool. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
LoRA SFT fine-tune on King CXXIV for native bookend emission (Phase 4).
All-time downloads
43
Public
Parameters
34.7B
71.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors71.9 GB · 100%
At a glance
- Model type
- qwen3_5_moe
- Access
- Public
- Created
- Sep 9, 2026
- Updated
- Sep 9, 2026
- SHA
- e387321d
Model card
From the Hugging Face model README
Albedo Qwen3.6-35B — Bookend Phase 4 SFT
LoRA SFT fine-tune on King CXXIV for native bookend emission (Phase 4).
Training
| Param | Value |
|---|---|
| Base | local_king/king_cxxiv (King CXXIV) |
| Data | data/sft/bookend_sft.jsonl (1000 rows, from bookend DPO chosen) |
| Epochs | 2 (32 steps) |
| LR | 1e-4 |
| LoRA r | 64 |
| Final adapter | checkpoints/bookend_phase4_sft/checkpoint-32 |
| Train loss | ~1.055 |
| mean_token_accuracy | ~90.7% |
Bundle layout
- SFT-tuned language weights (merged from checkpoint-32)
- Unchanged vision weights from King CXXIV
- Genesis metadata (validator
metadata_hashcompatible)
Built with offline merge_lora.py + shard repack per docs/FIELD_NOTES_TRAINING_AND_SUBMIT.md.
- Type
- qwen3_5_moe
- Created
- Sep 9, 2026
- Updated
- Sep 9, 2026