© 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. 55898
    Prev
    Next
    INDEX
    Explanations

    law of

    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
    å¸IJ篷
    -0.08
    unker
    -0.08
    çİ°è¡Į
    -0.08
    çĶľçĶľ
    -0.08
    æ°Ķ
    -0.08
    olec
    -0.08
    ression
    -0.08
    ãĥĪãĥ«
    -0.08
    çħŀ
    -0.08
    jh
    -0.08
    POSITIVE LOGITS
    é¦ĭ
    0.10
    ר×Ļ×Ŀ
    0.09
    endir
    0.09
    idth
    0.09
    å¾ĭ
    0.09
     Squad
    0.08
    åij½
    0.08
    åĴ½
    0.08
     Barnett
    0.08
     ÑĦинанÑģов
    0.08
    Activations Density 0.072%

    No Known Activations