© 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. 3313
    Prev
    Next
    INDEX
    Explanations

    announcing or publishing official information

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

    The neuron activates on reporting verbs in the past tense (e.g. “announced,” “revealed,” “recovered”).

    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
     혹은
    0.81
    .
    0.80
    !
    0.77
     וכ
    0.76
    }}.
    0.75
    ทุก
    0.73
     καὶ
    0.70
    ?
    0.69
     și
    0.67
     หรือ
    0.66
    POSITIVE LOGITS
     affidavits
    0.79
    ક્રિય
    0.75
     hostilities
    0.74
     पेशे
    0.72
     recounted
    0.71
    originally
    0.70
     unscathed
    0.69
     தனது
    0.68
    legends
    0.68
     recounting
    0.67
    Activations Density 0.010%

    No Known Activations