© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 31607
    Prev
    Next
    INDEX
    Explanations

    on surface

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    vor
    -0.10
    reb
    -0.09
    èĥ½è®©
    -0.09
     Rowe
    -0.09
    çĶ±ä¸ŃåĽ½
    -0.08
    opt
    -0.08
    iled
    -0.08
    Tank
    -0.08
    ä¹İ
    -0.08
    egrator
    -0.08
    POSITIVE LOGITS
    åľ°éĿ¢
    0.14
    表éĿ¢
    0.12
     повеÑĢÑħ
    0.11
    surface
    0.11
     surface
    0.11
    _codegen
    0.10
    çľ¼çļ®
    0.10
    ÑĤвеÑĢжда
    0.09
    urface
    0.09
    å¤§åľ°
    0.09
    Activations Density 0.074%

    No Known Activations