© 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. 54235
    Prev
    Next
    INDEX
    Explanations

    hole or holes

    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
    ians
    -0.10
    orst
    -0.09
    ively
    -0.09
    OUN
    -0.09
    ocom
    -0.09
    apos
    -0.09
    atel
    -0.09
    çļ¿
    -0.09
    ฯ
    -0.08
    à¹Į
    -0.08
    POSITIVE LOGITS
    holes
    0.12
    -hole
    0.11
    -eyed
    0.11
    hole
    0.11
    tober
    0.10
    ç©´
    0.10
    oader
    0.10
     зÑĢениÑı
    0.10
    busters
    0.09
    累累
    0.09
    Activations Density 0.026%

    No Known Activations