© 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. 71542
    Prev
    Next
    INDEX
    Explanations

    crow variations

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

    The neuron strongly activates whenever the token “Crow” (the character sequence c-r-o-w) appears, regardless of word context.

    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
    P
    -3.89
    S
    -3.84
    F
    -3.78
    G
    -3.73
    A
    -3.59
    E
    -3.56
    L
    -3.45
    M
    -3.41
    K
    -3.28
    T
    -3.22
    POSITIVE LOGITS
     questionSearch
    3.70
     GOtt
    3.36
    sukienka
    3.25
    an
    3.02
     丁目
    2.84
     放題
    2.83
    
    2.77
     Monfieur
    2.72
    ſtra
    2.67
     ſch
    2.66
    Activations Density 0.005%

    No Known Activations