© 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. 63282
    Prev
    Next
    INDEX
    Explanations

    scrambled or scrambling

    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
    imately
    -0.08
    atan
    -0.08
    ä¸Ĭåİ»
    -0.08
    ัà¸ĩ
    -0.08
    Adventure
    -0.08
    åĿIJ
    -0.08
    åĵ²
    -0.08
    .weights
    -0.08
    ITA
    -0.08
    ëĭ¨ì²´
    -0.08
    POSITIVE LOGITS
    ä¹¾åĿ¤
    0.10
    é©¿
    0.09
    atial
    0.09
     compreh
    0.09
    é£İæĥħ
    0.09
    éĢ¢
    0.09
    åı«å£°
    0.09
    iens
    0.08
    æīĵæŃ»
    0.08
    èĪĽ
    0.08
    Activations Density 0.002%

    No Known Activations