© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 124909
    Prev
    Next
    INDEX
    Explanations

    events and roles

    np_acts-logits-general · gemini-2.5-flash-lite

    routing to eccentric adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pharmaceut
    0.53
     counsellor
    0.46
     Drugs
    0.45
     बुखार
    0.45
     metabolic
    0.44
    "]').
    0.43
     drug
    0.43
     serum
    0.43
     therapeutics
    0.43
    ।’
    0.42
    POSITIVE LOGITS
    aré
    0.54
    reliant
    0.54
    ating
    0.52
    tere
    0.52
    م
    0.52
    Ex
    0.50
    ac
    0.50
    Change
    0.50
    Ac
    0.49
    planning
    0.49
    Activations Density 0.000%

    No Known Activations