TIE-Pilot
/

deepspec-drafter-ablations

speculative-decoding

Model card Files Files and versions

deepspec-drafter-ablations

32.3 GB

Ctrl+K

Ctrl+K

1 contributor

History: 17 commits

TIE-Pilot's picture

Remove internal note references from train_config.py

ffa48e2 verified 17 days ago

attnconv_b7_qwen3_4b
Remove internal note references from train_config.py 17 days ago
attnhead_b7_qwen3_4b
Remove internal note references from train_config.py 17 days ago
dflash_b7_qwen3_4b_1ep
Add dflash_b7_qwen3_4b_1ep @ step_2616 (1-epoch ablation arm) 21 days ago
dflash_b7_qwen3_4b_1ep_cemix_serve
Add dflash_b7_qwen3_4b_1ep_cemix_serve @ step_2616 (1-epoch ablation arm) 21 days ago
dflash_b7_qwen3_4b_1ep_tv
Add dflash_b7_qwen3_4b_1ep_tv @ step_2616 (1-epoch ablation arm) 21 days ago
dspark_b7_qwen3_4b_1ep
Add dspark_b7_qwen3_4b_1ep @ step_2616 (1-epoch ablation arm) 21 days ago
dspark_b7_qwen3_4b_1ep_shortconv
Add dspark_b7_qwen3_4b_1ep_shortconv @ step_2616 (1-epoch ablation arm) 21 days ago
official_dflash2_b8_qwen3_4b_1ep
Add official_dflash2_b8_qwen3_4b_1ep @ step_2616 (1-epoch ablation arm) 21 days ago
slotembed_b7_qwen3_4b
Add slotembed_b7_qwen3_4b @ step_2616 (1-epoch ablation arm) 21 days ago
v3a_condhead_b7_qwen3_4b
Add v3a_condhead_b7_qwen3_4b @ step_2616 (1-epoch ablation arm) 21 days ago
v3b_slotsplit_b7_qwen3_4b
Add v3b_slotsplit_b7_qwen3_4b @ step_2616 (1-epoch ablation arm) 21 days ago
xg_v21_b7_qwen3_4b
Add xg_v21_b7_qwen3_4b @ step_2616 (1-epoch ablation arm) 21 days ago
.gitattributes

1.52 kB
initial commit 21 days ago
README.md

5.7 kB
Use the official arXiv BibTeX entry 21 days ago