© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9222
    Prev
    Next
    INDEX
    Explanations

    effectiveness and performance

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

    The neuron fires strongly on words that describe system features, components, or processes—i.e. technical or functional terms in an instructional or descriptive context.

    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
    И
    1.10
    ्य
    1.03
    節目
    0.99
    ти
    0.98
    cS
    0.98
    cV
    0.97
    ин
    0.96
    gia
    0.95
     Dunia
    0.95
    {},
    0.94
    POSITIVE LOGITS
     flawlessly
    1.22
     kinks
    1.15
     작동
    1.10
    వంత
    1.06
    有效的
    1.04
     amended
    1.03
     clamps
    1.02
    सोई
    1.02
     efficacious
    1.02
     efficace
    1.01
    Activations Density 1.923%

    No Known Activations