© 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. 9311
    Prev
    Next
    INDEX
    Explanations

    "it's just" followed by descriptive words

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

    This neuron flags emphatic or evaluative expressions—particularly intensifiers and strong opinion markers (e.g. “just,” “plain as day,” “cruel”) that emphasize a statement.

    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
     delighted
    0.73
     excellent
    0.64
     pleased
    0.62
    にて
    0.59
    惊喜
    0.59
     newcomer
    0.59
    董事
    0.58
     (
    0.57
     photographs
    0.57
     unusual
    0.57
    POSITIVE LOGITS
     gotta
    0.96
     Gonna
    0.91
     俺
    0.85
     Están
    0.82
     Gotta
    0.81
     hypocrisy
    0.79
     bilhões
    0.79
    EndZ
    0.76
     cír
    0.75
     ríos
    0.73
    Activations Density 0.352%

    No Known Activations