© 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. 12132
    Prev
    Next
    INDEX
    Explanations

    mathematical and code structures"}, "response": "mathematical and code structures"}

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

    The neuron strongly activates on tokens that are part of numeric measurements or identifiers—especially floating-point or decimal numbers.

    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
    <0x80>
    0.75
     fond
    0.66
    ät
    0.64
    empt
    0.64
    ח
    0.63
     Tü
    0.62
     अरे
    0.61
    ת
    0.61
    band
    0.61
    ende
    0.60
    POSITIVE LOGITS
     radionuclide
    0.94
     aneur
    0.88
     warfarin
    0.86
     assistive
    0.84
    Missense
    0.82
     creatinine
    0.82
     calyce
    0.82
     cryogenic
    0.81
    OLYBDENUM
    0.80
     equaling
    0.80
    Activations Density 0.000%

    No Known Activations