© 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. 31-GEMMASCOPE-2-RES-16K
    7. 631
    Prev
    Next
    INDEX
    Explanations

    numbers and bullet points

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

    The neuron fires most strongly on uncommon or low‐frequency tokens (i.e. rare words).

    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
     www
    0.52
    ↵↵
    0.50
    www
    0.49
    .
    0.48
    based
    0.47
    ↵↵↵
    0.46
    )
    0.46
    ch
    0.46
    <b>
    0.45
    ’
    0.44
    POSITIVE LOGITS
     algéb
    0.66
     substring
    0.66
     usuário
    0.62
     tensor
    0.61
    ﷻ
    0.57
     drawable
    0.56
     deserial
    0.56
     fermion
    0.55
     rigidbody
    0.55
     modulo
    0.55
    Activations Density 0.000%

    No Known Activations