© 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. 48237
    Prev
    Next
    INDEX
    Explanations

    PillCam, pills, pillared

    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
    ettel
    -0.10
    ettle
    -0.10
    heed
    -0.10
    .annot
    -0.09
    shire
    -0.09
    beeld
    -0.09
    åijĺ
    -0.09
     اÙĦÙħسÙĦØŃØ©
    -0.09
    zeit
    -0.08
    pcf
    -0.08
    POSITIVE LOGITS
    åŃIJéĩĮ
    0.10
    ometer
    0.09
    omat
    0.09
    linger
    0.08
    ington
    0.08
    çĽĴ
    0.08
    Æ¡i
    0.08
    Ñģад
    0.08
     наÑģел
    0.08
    arity
    0.08
    Activations Density 0.005%

    No Known Activations