Downloads · 30 days
0
ace-step-sae-scores-mood
lukasz-staniszewski/ace-step-sae-scores-mood
ace-step-sae-scores-mood is a machine learning model from lukasz-staniszewski. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Per-timestep feature-importance score tables for the ACE-Step tf6/tf7 cross-attention SAEs: activations collected with the shared CONCEPTTOPROMPTS pairs (the same text every data-driven method in the benchmark uses) a…
Access
Public
Updated Aug 29, 2026
Repo size
16.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pkl5.5 MB · 100%
At a glance
- Access
- Public
- Created
- May 19, 2026
- Updated
- Aug 29, 2026
- SHA
- d44b1a9e
Model card
From the Hugging Face model README
ACE-Step SAE feature scores — mood
Per-timestep feature-importance score tables for the ACE-Step tf6/tf7 cross-attention
SAEs: activations collected with the shared CONCEPT_TO_PROMPTS pairs (the same text
every data-driven method in the benchmark uses) and encoded with the SAE in fp32.
Provenance (seed, git SHA) in the *_scores_config.json sidecars.
ctrl = SAESteeringController.from_pretrained(
"lukasz-staniszewski/ace-step-sae-scores-mood", hookpoint="transformer_blocks.7.cross_attn",
scores_cache_path="lukasz-staniszewski/ace-step-sae-scores-mood", score_filename="tf7_scores.pkl",
top_k=20, intervention_mode="steering_vector", alpha=10.0,
)
The original paper-era score tables remain available at revision 231f2763eded80426e3d68ba7db81f0ac63b8aef of this
repo (revision="231f2763eded80426e3d68ba7db81f0ac63b8aef" in hf_hub_download).
- Created
- May 19, 2026
- Updated
- Aug 29, 2026