© 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. 22-GEMMASCOPE-RES-131K
    7. 38505
    Prev
    Next
    INDEX
    Explanations

    legal documentation and citations

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

    The neuron strongly activates on numeric tokens and codes (digit sequences, hyphens, and number‐like strings).

    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
     tidigare
    -1.04
    dsquared
    -1.01
    madison
    -0.99
    kenstock
    -0.96
    lyst
    -0.94
     any
    -0.92
    ««
    -0.91
    adobe
    -0.91
     who
    -0.90
    kanak
    -0.90
    POSITIVE LOGITS
    ati
    1.06
    inn
    1.00
     chiese
    0.95
    既然
    0.93
    IB
    0.90
    ara
    0.90
    Pozri
    0.90
    ath
    0.89
    volent
    0.88
    建物
    0.87
    Activations Density 0.013%

    No Known Activations