© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108008
    Prev
    Next
    INDEX
    Explanations

    times and dates

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

    The neuron responds to numerical and enumerative tokens—digits, hyphens, date/time or section numbers—i.e. numeric literals and their punctuation in text.

    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
     đúng
    -0.86
     correcto
    -0.85
     хочет
    -0.84
     odası
    -0.82
     hút
    -0.82
    してくれた
    -0.81
     endfor
    -0.81
    бята
    -0.81
     maybe
    -0.81
     trend
    -0.79
    POSITIVE LOGITS
    )
    0.96
    ).
    0.89
    ヲ
    0.87
    stn
    0.85
     indispensable
    0.83
     Bux
    0.83
    indis
    0.82
    }
    0.81
     összefoglal
    0.81
    }\
    0.80
    Activations Density 0.005%

    No Known Activations