© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32310
    Prev
    Next
    INDEX
    Explanations

    Green functions or bundles

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

    The neuron fires strongly on standalone technical or domain‐specific keywords written in title‐case or all caps (e.g. “Green,” “ENGINE,” “ACTION_IMAGE_CAPTURE”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     regardé
    -0.80
    خدمات
    -0.74
    CGContext
    -0.71
    Ats
    -0.69
     overcrowding
    -0.68
     INR
    -0.68
     commandé
    -0.68
    陵
    -0.68
    熹
    -0.67
    مقد
    -0.66
    POSITIVE LOGITS
     Caesar
    0.70
    plans
    0.67
    чению
    0.66
    都
    0.66
     Distrito
    0.65
    prod
    0.64
    madas
    0.64
     COMPET
    0.63
     stř
    0.62
    detected
    0.61
    Activations Density 0.080%

    No Known Activations