© 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. 113785
    Prev
    Next
    INDEX
    Explanations

    yx, YX, Yv, YK, Yh, YM, YC, YH

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

    This neuron activates on and thus detects two‐letter alphabetic tokens (letter–letter bigrams).

    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
    there
    -3.22
    when
    -2.88
    e
    -2.70
    .......
    -2.66
     When
    -2.66
    滙
    -2.63
     There
    -2.55
    Kepala
    -2.55
    ８
    -2.45
    ９
    -2.45
    POSITIVE LOGITS
    
    2.83
    '
    2.81
    咈
    2.39
    撂
    2.34
    だったり
    2.34
     dando
    2.31
     be
    2.31
    回到了
    2.25
     solamente
    2.23
    匆
    2.22
    Activations Density 0.029%

    No Known Activations