© 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. 8568
    Prev
    Next
    INDEX
    Explanations

    recognizing and focusing

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

    The neuron strongly responds to verbs in their present‐participle/gerund form (i.e. “-ing” action 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
    !’
    0.81
    )’
    0.81
    ’)
    0.80
    ’
    0.79
    ’,
    0.75
    APPING
    0.75
    ﻒ
    0.72
    DBES
    0.71
    ’;
    0.71
    _{
    0.70
    POSITIVE LOGITS
    他
    0.96
     лишь
    0.86
    或許
    0.84
     jusqu
    0.83
    或许
    0.83
     it
    0.81
    то
    0.81
     siitä
    0.81
    lerle
    0.81
     лише
    0.80
    Activations Density 0.399%

    No Known Activations