© 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. 23461
    Prev
    Next
    INDEX
    Explanations

    velocomotion, velour, velocity

    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
    arak
    -0.10
    ness
    -0.10
    ÑģÑĤвеннÑĭе
    -0.09
    places
    -0.09
    olina
    -0.09
    媳
    -0.09
    æĢ§å¼º
    -0.09
    ../../
    -0.09
     Tits
    -0.09
    yll
    -0.08
    POSITIVE LOGITS
    0.12
    ocities
    0.12
    _CB
    0.09
    ailles
    0.09
     Ñģобой
    0.09
    ozo
    0.09
    oy
    0.09
    ower
    0.09
     ë¶Ī구íķĺê³ł
    0.09
    ÑĥÑī
    0.09
    Activations Density 0.014%

    No Known Activations