© 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. 313
    Prev
    Next
    INDEX
    Explanations

    988 and 741741

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

    This neuron specializes in detecting long sequences of numeric characters (e.g., phone numbers or multi-digit IDs).

    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
     prerog
    0.23
    ₁,
    0.23
     récit
    0.23
    <unused5>
    0.22
     oeuvre
    0.21
     kategor
    0.21
     paradigma
    0.21
    ¹,
    0.20
    <unused15>
    0.20
     affections
    0.20
    POSITIVE LOGITS
    -
    0.36
    0
    0.32
    2
    0.30
    9
    0.30
    8
    0.29
    4
    0.27
    3
    0.27
    The
    0.26
    L
    0.26
    A
    0.26
    Activations Density 0.418%

    No Known Activations