© 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. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 46924
    Prev
    Next
    INDEX
    Explanations

    protective patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    bcywinski/Olmo-3-32B-Base-SAE/saes_allenai_Olmo-3-1125-32B_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
     âĻ¥
    -0.11
     incidental
    -0.11
     openness
    -0.11
     graz
    -0.10
     fleeting
    -0.10
     âĿ¤
    -0.10
     confidence
    -0.10
     jealousy
    -0.10
     Countdown
    -0.10
     Corm
    -0.10
    POSITIVE LOGITS
     subsystem
    0.14
     dysfunctional
    0.12
     dysfunction
    0.12
     perfection
    0.11
     Dysfunction
    0.11
     alcoholic
    0.10
     alcohol
    0.10
     Roles
    0.10
     addicts
    0.10
    Roles
    0.10
    Activations Density 0.043%

    No Known Activations