© 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. 24-GEMMASCOPE-2-RES-16K
    7. 765
    Prev
    Next
    INDEX
    Explanations

    compliance and transfer context

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

    The neuron fires on multi‐word, title‐style noun phrases (section headings) set off in Title Case.

    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
    Estoy
    0.50
     장면
    0.49
    ឬ
    0.45
    白い
    0.44
    紺
    0.44
    BRO
    0.43
    Bạn
    0.43
    鶚
    0.43
     жовт
    0.42
    También
    0.42
    POSITIVE LOGITS
     maximization
    0.63
     inherent
    0.61
     filter
    0.59
     filtering
    0.57
     imbalances
    0.56
     ratios
    0.56
     imbalance
    0.56
     impedance
    0.55
     ratio
    0.55
     reallocation
    0.55
    Activations Density 0.000%

    No Known Activations