© 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. 22-GEMMASCOPE-RES-131K
    7. 71139
    Prev
    Next
    INDEX
    Explanations

    unknown: failed to open stream

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

    The neuron is specialized to spot programming‐style tokens—particularly class and API identifiers (like UI, NS, UIActivityIndicatorView) and method‐call syntax (dots, colons, parentheses) in 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_22/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
     április
    -1.20
    every
    -1.09
    despite
    -1.08
     acclaimed
    -1.06
     ซึ่ง
    -1.05
    six
    -1.05
    usst
    -1.03
     spécialiste
    -1.02
    three
    -1.02
    two
    -1.02
    POSITIVE LOGITS
     strangely
    1.25
    عاون
    1.23
     what
    1.21
     specially
    1.21
     any
    1.20
     that
    1.16
     blatantly
    1.14
     one
    1.13
     navigating
    1.13
     horribly
    1.12
    Activations Density 0.129%

    No Known Activations