© 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. 55237
    Prev
    Next
    INDEX
    Explanations

    contextual request or context type

    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
    ниÑĨа
    -0.10
    --------------------------------
    -0.09
    inka
    -0.09
    cxx
    -0.09
    Tro
    -0.09
    ard
    -0.09
    å°Ĩé¢Ĩ
    -0.09
    æ¦ľ
    -0.08
    çŁ³åĮĸ
    -0.08
    thest
    -0.08
    POSITIVE LOGITS
    ually
    0.15
    ual
    0.14
    ä¸įæ¸ħ
    0.11
    /context
    0.11
    MenuStrip
    0.11
    \Context
    0.10
     CONTEXT
    0.10
    åįģåĪĨéĩįè¦ģ
    0.09
     ctx
    0.09
    å¾Īæ·±
    0.09
    Activations Density 0.049%

    No Known Activations