© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 39522
    Prev
    Next
    INDEX
    Explanations

    id fields

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

    The neuron fires on numeric identifier tokens (numbers used as IDs in code or query text).

    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
    _",
    -0.91
    able
    -0.90
     why
    -0.88
    ACHUSETTS
    -0.88
     pesso
    -0.87
     可愛
    -0.86
     ᾿
    -0.85
    <h2>
    -0.85
     regista
    -0.84
     可愛い
    -0.83
    POSITIVE LOGITS
    钊
    0.94
     zostać
    0.93
    
    0.92
    に伴
    0.91
     altid
    0.90
    moires
    0.88
     muž
    0.88
    じゃないですか
    0.88
    ürtel
    0.87
     cambian
    0.87
    Activations Density 0.014%

    No Known Activations