© 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. 19182
    Prev
    Next
    INDEX
    Explanations

    possible values or ways

    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
    èĭ¦èĭ¦
    -0.10
    è¯ķåĽ¾
    -0.10
    yard
    -0.09
    GMT
    -0.09
    alus
    -0.09
     possibility
    -0.09
    ney
    -0.09
    ä¿©
    -0.08
    Ctx
    -0.08
    ŀæĢ§
    -0.08
    POSITIVE LOGITS
    ities
    0.13
    keiten
    0.12
    berra
    0.10
    orate
    0.10
    ITY
    0.10
    /legal
    0.09
    ined
    0.09
    æ®Ĩ
    0.09
    alker
    0.09
    éĻIJãĤĬ
    0.09
    Activations Density 0.049%

    No Known Activations