© 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. 22-GEMMASCOPE-RES-131K
    7. 66671
    Prev
    Next
    INDEX
    Explanations

    code logic

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

    The neuron fires strongly on subwords of camelCase identifiers, i.e. the individual “words” (like “one,” “Time,” “Task,” etc.) that make up a camel-cased name.

    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
    OPENING
    -0.82
    Viol
    -0.81
    ulz
    -0.81
    AFFIRMED
    -0.78
    ี่ยว
    -0.78
     mire
    -0.77
    átula
    -0.76
    ností
    -0.76
    znamu
    -0.75
     około
    -0.73
    POSITIVE LOGITS
    ﻷ
    1.00
    uruh
    0.87
    atea
    0.85
    viewDidLoad
    0.84
     Reset
    0.82
    érience
    0.82
    ?></
    0.79
    ->$
    0.78
    reset
    0.77
     Unsupported
    0.77
    Activations Density 0.002%

    No Known Activations