© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9080
    Prev
    Next
    INDEX
    Explanations

    connecting independent clauses with and

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

    The neuron strongly activates on numeric tokens (especially multi-digit numbers or decimals).

    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
     অর্থাৎ
    0.70
    $.)
    0.61
    অর্থাৎ
    0.61
     يعني
    0.60
     அதாவது
    0.57
     즉
    0.56
    យៈ
    0.55
     یعنی
    0.55
    OCI
    0.54
    Bu
    0.54
    POSITIVE LOGITS
    romeda
    0.86
    ERSON
    0.72
    rews
    0.70
    erson
    0.70
    rogens
    0.70
    ersen
    0.64
    rà
    0.64
     then
    0.62
     whatnot
    0.61
    谒
    0.61
    Activations Density 0.075%

    No Known Activations