© 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. 84097
    Prev
    Next
    INDEX
    Explanations

    down, below, or a follow-up

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

    The neuron strongly activates on CJK (Chinese/Japanese/Korean) ideograph 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
    𖥧
    -2.83
    衎
    -2.39
    衚
    -2.27
     また
    -2.23
    F
    -2.20
    ༵
    -2.17
    P
    -2.16
     populi
    -2.13
    T
    -2.09
    ;
    -2.09
    POSITIVE LOGITS
    顿时
    2.83
    鸝
    2.75
    饜
    2.58
     với
    2.56
    神
    2.56
     עם
    2.55
    ↵↵
    2.55
    9
    2.53
     penchant
    2.50
     myriad
    2.47
    Activations Density 0.000%

    No Known Activations