© 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
    3. Gemma-3-4B-IT
    5. 17-GEMMASCOPE-2-RES-16K
    7. 2734
    Prev
    Next
    INDEX
    Explanations

    purposes therein, year, crisis, shall seem

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_17_width_16k_l0_medium
    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
    ,
    0.77
    <0xC2>
    0.75
     let
    0.75
    ↵
    0.75
     right
    0.71
    0.70
     can
    0.69
     get
    0.68
     take
    0.68
     might
    0.65
    POSITIVE LOGITS
    ⵃ
    1.16
    organizzazione
    1.13
    ⵕ
    1.11
    dinitro
    1.10
    evoluzione
    1.10
    ignées
    1.09
    avasena
    1.08
    assapi
    1.08
    posición
    1.08
    ziła
    1.06
    Activations Density 0.006%

    No Known Activations