© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72140
    Prev
    Next
    INDEX
    Explanations

    initial or once

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    annulation
    -0.84
     koła
    -0.82
    ientemente
    -0.80
     nécessité
    -0.76
     abbildung
    -0.76
     przypad
    -0.76
    ői
    -0.75
     раньше
    -0.75
    XP
    -0.74
     mellom
    -0.74
    POSITIVE LOGITS
     initial
    2.20
    initial
    1.80
     inicial
    1.73
    初始
    1.67
    Initial
    1.64
    inicial
    1.45
     initiale
    1.43
     Initial
    1.38
     once
    1.38
    INITIAL
    1.37
    Activations Density 0.153%

    No Known Activations