© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50662
    Prev
    Next
    INDEX
    Explanations

    August dates

    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
    erah
    -0.11
    ìľ¨
    -0.10
    å®īå®ļ
    -0.10
    lessly
    -0.09
    arning
    -0.09
    clud
    -0.09
    _ADMIN
    -0.09
    æ³
    -0.09
    ction
    -0.09
    haul
    -0.09
    POSITIVE LOGITS
    ively
    0.12
    sWith
    0.11
    eenth
    0.11
    eer
    0.11
    ters
    0.10
    mar
    0.10
    imuth
    0.09
    ouch
    0.09
    ments
    0.09
    elaide
    0.09
    Activations Density 0.037%

    No Known Activations