© 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. 12507
    Prev
    Next
    INDEX
    Explanations

    mouse and rat

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

    The neuron is detecting numeric tokens (numbers or digit‐containing strings) in the text.

    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
    音声
    0.61
    绿
    0.60
    واصل
    0.57
    alte
    0.56
    lamps
    0.54
     Diagnose
    0.53
     Mer
    0.52
     Acces
    0.52
     cáps
    0.51
    empt
    0.50
    POSITIVE LOGITS
     rat
    1.30
     rodent
    1.20
     ratification
    1.06
    Rat
    1.05
     Rat
    1.03
     rodents
    1.02
    鼠
    0.98
     ratified
    0.96
     rats
    0.94
    🐀
    0.94
    Activations Density 0.076%

    No Known Activations