© 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. 65696
    Prev
    Next
    INDEX
    Explanations

    actions, outcomes, or academic steps

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

    The neuron strongly activates on low‐frequency or otherwise unusual tokens (e.g. rare words, multi‐dot ellipses, or uncommon terms) regardless of language.

    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
    いくつか
    -2.56
    囫
    -2.42
     einigen
    -2.42
     jenen
    -2.28
     Hinblick
    -2.25
    ,
    -2.25
     vielen
    -2.19
     этим
    -2.17
    婀
    -2.14
    妧
    -2.13
    POSITIVE LOGITS
     detección
    2.56
    狁
    2.50
    </em>
    2.47
    ”，
    2.38
    <td>
    2.30
     intervención
    2.14
    使得
    2.03
     墊
    2.02
    ")
    1.98
    
    1.98
    Activations Density 0.000%

    No Known Activations