© 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. Gemma-3-27B-IT
    5. 34-GEMMASCOPE-2-TRANSCODER-262K
    7. 17380
    Prev
    Next
    INDEX
    Explanations

    logical and sensible reasoning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_34_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    barrels
    0.36
     कदा
    0.35
    количество
    0.34
    ֜
    0.33
     confronted
    0.32
     услы
    0.32
    کنندگان
    0.32
    阳
    0.32
    Characteristic
    0.32
     виправи
    0.32
    POSITIVE LOGITS
     reasonable
    1.01
    reasonable
    0.87
     Reasonable
    0.85
    合理
    0.76
     logical
    0.74
     reasonableness
    0.73
     разум
    0.72
     sensible
    0.71
     basic
    0.66
    合理的
    0.65
    Activations Density 0.140%

    No Known Activations