© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2055
    Prev
    Next
    INDEX
    Explanations

    code blocks and explanations

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

    The neuron consistently fires on numeric literal tokens (including integers and floating‐point 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_31_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
    Sixth
    0.91
    Civil
    0.84
    Seventh
    0.82
    Fifth
    0.81
    Div
    0.80
    Third
    0.78
    Person
    0.76
    6
    0.76
    3
    0.75
    seventh
    0.74
    POSITIVE LOGITS
     ∠
    1.11
     $\
    1.03
     `$
    1.00
     denoted
    0.98
     "".
    0.98
     $-\
    0.97
     analyte
    0.97
     $\{
    0.95
     både
    0.95
     {}".
    0.94
    Activations Density 0.171%

    No Known Activations