© 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. 10-GEMMASCOPE-RES-131K
    4. 104245
    Prev
    Next
    INDEX
    Explanations

    identification

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

    This neuron fires on the initial subword of capitalized proper names (e.g. "Prim" in Primrose or "Aval" in Avalanche).

    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
    ↵
    -3.20
    }
    -3.09
    '
    -2.67
    </h5>
    -2.66
    $
    -2.53
     were
    -2.47
    ve
    -2.22
    ”
    -2.17
    ва
    -2.16
     ∼
    -2.16
    POSITIVE LOGITS
    ﹍﹍
    2.80
     aktivi
    2.61
     sprüche
    2.56
    
    2.44
     жк
    2.42
     gewor
    2.41
     hintergrund
    2.41
     تجه
    2.38
    玏
    2.23
     Descri
    2.22
    Activations Density 0.002%

    No Known Activations