© 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. 48410
    Prev
    Next
    INDEX
    Explanations

    hole and followed by specific words

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

    The neuron fires on any occurrence of “hole” or “holes” (including as part of compound words).

    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
    涠
    -2.47
    
    -2.42
    剿
    -2.38
    艽
    -2.27
     descobri
    -2.20
     are
    -2.19
    痾
    -2.17
    ykjav
    -2.16
    enablog
    -2.14
     pôr
    -2.14
    POSITIVE LOGITS
    3
    3.17
    0
    2.80
    N
    2.59
    6
    2.56
    1
    2.56
    “
    2.41
    5
    2.39
    7
    2.36
    ２
    2.31
    W
    2.28
    Activations Density 0.006%

    No Known Activations