© 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. 12285
    Prev
    Next
    INDEX
    Explanations

    in python

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

    The main thing this neuron does is detect numeric tokens (e.g. years, measurements, or other standalone 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
    ในปี
    0.86
    Người
    0.78
    j
    0.75
     вз
    0.74
    Все
    0.73
     внутрен
    0.72
    И
    0.70
    ம்
    0.69
    frey
    0.69
    decade
    0.69
    POSITIVE LOGITS
     accordance
    1.85
     lieu
    1.49
     vacuo
    1.47
     vitro
    1.46
     conjunction
    1.45
     perpetuity
    1.38
    volving
    1.36
     relation
    1.34
     terms
    1.33
     toto
    1.27
    Activations Density 0.212%

    No Known Activations