© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 12170
    Prev
    Next
    INDEX
    Explanations

    words indicate timing or sequence

    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
     both
    -1.27
     what
    -1.26
    μένα
    -1.14
    cargas
    -1.13
     any
    -1.12
    と言えば
    -1.09
     пыта
    -1.09
     mejorar
    -1.09
    ständige
    -1.08
    furl
    -1.07
    POSITIVE LOGITS
     only
    1.62
     只
    1.37
    complished
    1.30
     last
    1.27
     early
    1.18
     after
    1.16
     until
    1.16
     فقط
    1.15
     starting
    1.14
     AQUÍ
    1.14
    Activations Density 0.070%

    No Known Activations