© 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. 6801
    Prev
    Next
    INDEX
    Explanations

    numbers following 5 or 9

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

    This neuron detects numeric tokens, especially multi‐digit sequences such as phone numbers or other numeric identifiers.

    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
     thorax
    0.52
     Lo
    0.51
     lus
    0.51
     supervis
    0.50
    ตัว
    0.49
     দূতাব
    0.48
     basically
    0.47
    책
    0.47
     দূতাবাস
    0.47
     accused
    0.46
    POSITIVE LOGITS
    岐阜
    0.53
     ది
    0.50
     एवरेज
    0.50
    зовая
    0.50
     cumplen
    0.49
    හෙ
    0.49
    omicide
    0.49
    ייה
    0.48
    aniach
    0.48
     liegen
    0.47
    Activations Density 0.159%

    No Known Activations