© 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. 24-GEMMASCOPE-2-RES-16K
    7. 6983
    Prev
    Next
    INDEX
    Explanations

    positively descriptive language

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

    The neuron is sensitive to biographical event words—particularly life‐event verbs like “born,” “died,” “married,” etc., often marking dates.

    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
     আমায়
    0.59
    षे
    0.58
    kowej
    0.57
    лым
    0.56
     Север
    0.56
    atering
    0.55
    멈
    0.55
    ęb
    0.54
    amantha
    0.54
     اذ
    0.54
    POSITIVE LOGITS
     goodies
    0.65
     lousy
    0.65
     heinous
    0.63
     innocent
    0.58
     bogus
    0.58
     mirac
    0.57
     dubious
    0.57
     sickly
    0.57
     idiosync
    0.57
     micrographs
    0.57
    Activations Density 0.025%

    No Known Activations