© 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. 34-GEMMASCOPE-RES-131K
    7. 116367
    Prev
    Next
    INDEX
    Explanations

    method parameter names

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

    The neuron fires on programming‐language type tokens and other formal code identifiers (e.g. Java types like CharSequence, int, MouseEvent, Bundle, etc.).

    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
     obok
    -0.91
     taşlı
    -0.90
    kapat
    -0.88
     tästä
    -0.88
    [.
    -0.87
     SITY
    -0.86
     UNIVE
    -0.86
     lateinit
    -0.85
     mild
    -0.85
    zhen
    -0.85
    POSITIVE LOGITS
     arg
    2.03
    arg
    1.64
     Arg
    1.34
    Arg
    1.32
     args
    1.27
    ARG
    1.15
     ARG
    1.14
    Args
    0.98
     still
    0.95
     Args
    0.93
    Activations Density 0.014%

    No Known Activations