© 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. 10-GEMMASCOPE-RES-131K
    7. 44296
    Prev
    Next
    INDEX
    Explanations

    list and object manipulation

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

    The neuron fires on identifiers that are being called as functions or methods (e.g. tokens like “append”, “range”, “insert”, “repeat”, “send” followed by parentheses).

    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
     “
    -2.38
    ቭ
    -2.19
    鬟
    -2.17
    ቬ
    -1.88
    ’
    -1.85
     perhaps
    -1.84
    ’:
    -1.84
     \
    -1.84
     “…
    -1.84
    ’，
    -1.80
    POSITIVE LOGITS
    
    1.63
    5
    1.63
    Donativos
    1.60
     員
    1.59
    karena
    1.58
    𖧷
    1.57
    the
    1.54
    Nummer
    1.52
     مسلمان
    1.52
    Adding
    1.49
    Activations Density 0.001%

    No Known Activations