© 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. 21705
    Prev
    Next
    INDEX
    Explanations

    importance

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

    The neuron activates on the adjective meaning “important” (e.g. English “important,” Russian “важно,” Latvian “svarīg,” etc.).

    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
    蝙
    -2.38
    ︲
    -2.19
    iſ
    -2.09
    蟊
    -2.03
    𐤔
    -1.97
    iyle
    -1.94
     bonitos
    -1.92
    
    -1.87
    雰
    -1.87
    OLOGIE
    -1.87
    POSITIVE LOGITS
     важно
    1.99
    and
    1.97
    X
    1.96
    始め
    1.91
    or
    1.88
    ほとんど
    1.84
     they
    1.80
    ED
    1.80
    but
    1.77
    ATION
    1.77
    Activations Density 0.004%

    No Known Activations