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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11707
    Prev
    Next
    INDEX
    Explanations

    This neuron primarily fires on numeric or number‐like tokens (especially floating‐point numerals).

    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_24_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
    .
    0.48
     perquè
    0.42
     или
    0.42
     alebo
    0.42
    )
    0.39
    ).
    0.38
    _
    0.38
    ))
    0.37
     porque
    0.37
    ().
    0.37
    POSITIVE LOGITS
     sumptuous
    0.46
     eclectic
    0.43
     delectable
    0.42
     unassuming
    0.39
     opulent
    0.39
     pleasurable
    0.39
     exquisitely
    0.39
     exhilarating
    0.38
     impeccable
    0.38
     astounding
    0.37
    Activations Density 0.000%

    No Known Activations