© 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. 22633
    Prev
    Next
    INDEX
    Explanations

    store location and details

    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
    ote
    -0.11
    åįģ足
    -0.11
    uate
    -0.10
    /her
    -0.10
    /from
    -0.10
    à¹Į
    -0.09
    sert
    -0.09
    ArrayOf
    -0.09
    erator
    -0.09
    çĦ¶æĺ¯
    -0.09
    POSITIVE LOGITS
    OrUpdate
    0.13
    house
    0.12
    keeper
    0.12
    front
    0.12
    keepers
    0.11
    ability
    0.11
    pping
    0.10
    ings
    0.10
    liness
    0.10
    ucción
    0.10
    Activations Density 0.049%

    No Known Activations