© 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. 2360
    Prev
    Next
    INDEX
    Explanations

    describing nature, structure, or function

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

    The neuron fires strongly on abstract, nominalized concept words (e.g. “situation,” “structure,” “content,” etc.).

    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
     που
    0.52
     yang
    0.46
    Te
    0.46
     เกิด
    0.46
     Whose
    0.46
    the
    0.44
     Ç
    0.44
     که
    0.44
    Berger
    0.44
    Which
    0.43
    POSITIVE LOGITS
     adik
    0.52
     fluctuate
    0.51
     respectivas
    0.50
    ﻴ
    0.49
    桡
    0.48
    цы
    0.47
     выпа
    0.47
     ilişkin
    0.47
    łą
    0.46
    blackberry
    0.46
    Activations Density 0.001%

    No Known Activations