© 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. 22-GEMMASCOPE-RES-131K
    7. 26325
    Prev
    Next
    INDEX
    Explanations

    not listed or repeated

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

    The neuron detects numeric tokens (digits and numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Flashback
    -0.86
     both
    -0.85
     even
    -0.84
    詭
    -0.84
     chiếc
    -0.83
     impartial
    -0.83
     Regardless
    -0.80
     причем
    -0.77
     whatever
    -0.77
     Whatever
    -0.75
    POSITIVE LOGITS
     repetitive
    1.13
    节省
    1.10
    Instead
    1.09
    不再
    1.07
    重复
    1.07
     unnecessary
    1.05
     repeated
    1.02
     repeti
    1.02
     instead
    1.01
     Instead
    1.01
    Activations Density 0.091%

    No Known Activations