© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113573
    Prev
    Next
    INDEX
    Explanations

    Sonny, waterfront, нет

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

    The neuron flags whichever token is the most frequently repeated term in a given text passage.

    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
    </i>
    -3.80
    <
    -2.98
    3
    -2.98
    With
    -2.91
    5
    -2.86
    <td>
    -2.80
    As
    -2.73
    So
    -2.67
    Since
    -2.66
    Initially
    -2.55
    POSITIVE LOGITS
    a
    2.80
    媜
    2.75
     cilindros
    2.69
     was
    2.66
    謦
    2.53
    phism
    2.50
    檍
    2.45
    🫅
    2.45
    🪆
    2.45
    幺
    2.45
    Activations Density 0.004%

    No Known Activations