© 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. 63605
    Prev
    Next
    INDEX
    Explanations

    fringe areas, fringe players, fringe of the Sahel

    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
    iation
    -0.09
    æĹ¶è¾°
    -0.09
    liness
    -0.09
    æł¸å¿ĥåĮº
    -0.09
     carne
    -0.08
    anes
    -0.08
    nown
    -0.08
    enus
    -0.08
    son
    -0.08
    oons
    -0.08
    POSITIVE LOGITS
    adier
    0.12
    ìĤ¬íķŃ
    0.11
    olis
    0.11
    nier
    0.10
    èī²å½©
    0.10
    gle
    0.10
    åŃIJéĩĮ
    0.10
    åľ°å¸¦
    0.10
    activ
    0.09
    füg
    0.09
    Activations Density 0.002%

    No Known Activations