© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10831
    Prev
    Next
    INDEX
    Explanations

    explicit content or code errors

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

    The neuron primarily detects floating‐point numbers (decimal values) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WorkingDirectory
    0.99
     asociaciones
    0.90
     expedient
    0.87
     remedi
    0.87
    шенный
    0.86
    寔
    0.86
     patrones
    0.86
     pInBuffer
    0.85
     dieting
    0.85
     dexterity
    0.85
    POSITIVE LOGITS
    https
    0.80
    terraform
    0.74
    Der
    0.73
    ક
    0.70
    marine
    0.68
    ސ
    0.67
     encore
    0.66
    fi
    0.66
     de
    0.64
    ado
    0.64
    Activations Density 0.000%

    No Known Activations