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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66397
    Prev
    Next
    INDEX
    Explanations

    Friday Night Lights

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

    The neuron is sensitive to technical or “code‐style” keywords and identifiers (things like macro parameters, annotations, template declarations, include headers, function/type names, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    しか
    -0.77
     puffed
    -0.76
     MaterialApp
    -0.74
    rfloor
    -0.73
    pendi
    -0.72
    來
    -0.71
    conse
    -0.71
     непре
    -0.71
    Conse
    -0.71
     его
    -0.70
    POSITIVE LOGITS
    cja
    0.89
    держка
    0.88
     avoient
    0.88
    DNEY
    0.85
    rierung
    0.83
     forKey
    0.82
     iveco
    0.81
    tazione
    0.80
     LXXX
    0.79
     comenzado
    0.79
    Activations Density 0.027%

    No Known Activations