© 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. 34-GEMMASCOPE-RES-131K
    7. 34745
    Prev
    Next
    INDEX
    Explanations

    ctypes, Clearing, Here

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

    This neuron fires on the whitespace indentation and structural punctuation (braces, parentheses, etc.) used to format code.

    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
     wastewater
    -0.80
     salivary
    -0.75
     coercion
    -0.74
    mutate
    -0.74
     reimbursement
    -0.74
     rookie
    -0.73
     kỷ
    -0.73
     Visit
    -0.73
     Ortsteil
    -0.72
    ɵɵ
    -0.72
    POSITIVE LOGITS
     GAIN
    0.87
    persona
    0.86
    Hvor
    0.84
    Hvem
    0.84
     fuff
    0.84
     قوان
    0.82
    Molto
    0.82
     环境
    0.80
    Abbiamo
    0.80
     Modul
    0.79
    Activations Density 0.006%

    No Known Activations