© 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. 10-GEMMASCOPE-RES-131K
    7. 41224
    Prev
    Next
    INDEX
    Explanations

    clock times

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

    The neuron is specialized to pick out numeric tokens within dates and timestamps (e.g. hours, minutes, seconds, years).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    1
    -2.06
    .'
    -1.93
    
    -1.79
    .<
    -1.76
    .
    -1.67
    棡
    -1.63
    .\\
    -1.62
    .“
    -1.59
    .?
    -1.58
    .”
    -1.57
    POSITIVE LOGITS
    ಞ
    2.23
    谵
    1.91
     ſou
    1.84
    you
    1.76
    rrggbb
    1.73
    ד
    1.70
     originates
    1.70
     cérami
    1.66
    ჯ
    1.66
     intend
    1.63
    Activations Density 0.039%

    No Known Activations