© 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. 14102
    Prev
    Next
    INDEX
    Explanations

    model responses

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

    This neuron is highly sensitive to numeric tokens—i.e. any kind of number or code (including question/answer numbers, probabilities, dollar amounts, etc.).

    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
    ort
    0.67
    య
    0.67
    ange
    0.62
    。[
    0.62
    <h2>
    0.60
    <0xBA>
    0.58
    argument
    0.58
    ర
    0.57
    ALI
    0.56
    ans
    0.55
    POSITIVE LOGITS
     Norwalk
    0.67
     보도록
    0.66
     Wexford
    0.66
     Voyage
    0.65
     resided
    0.65
     काही
    0.65
     Monique
    0.65
     Resonance
    0.64
     استخدم
    0.64
     resonated
    0.64
    Activations Density 0.000%

    No Known Activations