© 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. 3-GEMMASCOPE-2-TRANSCODER-262K
    7. 122604
    Prev
    Next
    INDEX
    Explanations

    A bit more

    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_3_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
    i
    0.91
    you
    0.89
    was
    0.87
    9
    0.87
    iem
    0.84
    ires
    0.82
    ra
    0.82
    ores
    0.79
    a
    0.79
    are
    0.78
    POSITIVE LOGITS
    in
    0.91
    ח
    0.79
    湡
    0.79
     bagno
    0.75
    ல்
    0.73
     percorso
    0.73
     dicembre
    0.70
    গ
    0.69
    بری
    0.68
     chauss
    0.68
    Activations Density 0.077%

    No Known Activations