© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 7977
    Prev
    Next
    INDEX
    Explanations

    image framing or descriptions

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

    The neuron primarily fires on tokens representing numeric values, measurements, prices, and other number‐like strings.

    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
     jefe
    0.65
    از
    0.65
    ↵
    0.63
    '
    0.62
    1
    0.61
     M
    0.60
    ็
    0.60
     A
    0.59
    ənd
    0.59
    Ü
    0.59
    POSITIVE LOGITS
    greencar
    0.81
     വാഹന
    0.72
    dagen
    0.72
    administrat
    0.72
    FirebaseAuth
    0.70
    thisStudent
    0.69
    quanto
    0.68
    觓
    0.68
    <unused4>
    0.67
    <unused45>
    0.67
    Activations Density 0.062%

    No Known Activations