© 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. 9124
    Prev
    Next
    INDEX
    Explanations

    party, whole, assignment, Fed

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

    The neuron strongly activates on open-class “content” tokens (longer, information-bearing words like healed, research, writer, appreciate, raised, grade, sale, etc.) rather than on short function 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_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
    나
    0.99
    an
    0.97
    К
    0.95
    entric
    0.91
    Мо
    0.90
    se
    0.89
    ak
    0.86
    세
    0.85
    z
    0.84
    라
    0.84
    POSITIVE LOGITS
    тились
    0.91
     screws
    0.89
    туры
    0.86
     chestnuts
    0.86
     twigs
    0.86
     tassels
    0.86
     turnips
    0.85
     giraffe
    0.84
     avocados
    0.82
     cucumbers
    0.82
    Activations Density 0.000%

    No Known Activations