william-0g
/

MSA-Qwen3.5

memory-sparse-attention

generative-retrieval

Model card Files Files and versions

316 kB

Ctrl+K

Ctrl+K

1 contributor

History: 3 commits

william-0g's picture

Add two-stage SFT: S1/S2 trainer, interleave dataset, launch script, method doc + S2 OOM record

c154876 verified 24 days ago

code
Add MSA x Qwen3.5 CPT trainer, model overlay, and training docs 25 days ago
docs
Add MSA x Qwen3.5 CPT trainer, model overlay, and training docs 25 days ago
sft
Add two-stage SFT: S1/S2 trainer, interleave dataset, launch script, method doc + S2 OOM record 24 days ago
.gitattributes

1.52 kB
initial commit 25 days ago
README.md

3.28 kB
Add two-stage SFT: S1/S2 trainer, interleave dataset, launch script, method doc + S2 OOM record 24 days ago