© 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. 10-GEMMASCOPE-RES-131K
    7. 119510
    Prev
    Next
    INDEX
    Explanations

    program date and status codes

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

    Looking at the activations, this neuron fires strongly on: - "fave" (64, 54.25) - "NSDate" (59.25, 53.75, 57.75) - "Renegade" (58.75

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron strongly activates on programming‐style identifiers and technical terms (e.g. CamelCase class names and annotation tokens).

    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
    </h3>
    -4.63
    u
    -3.97
    s
    -3.13
    0
    -2.92
    '
    -2.73
    </strong>
    -2.72
    as
    -2.70
    many
    -2.59
     :}
    -2.45
    n
    -2.41
    POSITIVE LOGITS
    </b>
    2.66
    
    2.30
     Then
    2.20
     大会
    2.16
    orteur
    2.13
    ของ
    2.11
    หรือ
    2.09
    啭
    2.03
    骘
    2.03
     ако
    2.02
    Activations Density 0.033%

    No Known Activations