© 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. 773
    Prev
    Next
    INDEX
    Explanations

    in the

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

    This neuron activates on multi‐token proper names or titles—especially organization or product names rendered as consecutive capitalized 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
     محاضرت
    0.43
    illons
    0.42
     coasters
    0.42
     ridicu
    0.42
    ल्लाला
    0.42
    𒃶
    0.42
    퐶
    0.42
    hiqdev
    0.41
    糇
    0.41
    vaient
    0.41
    POSITIVE LOGITS
    _
    0.61
    -
    0.42
    ↵
    0.38
    \_
    0.38
    ↵↵
    0.36
    2
    0.36
    /
    0.35
     Young
    0.35
    )
    0.33
    9
    0.33
    Activations Density 0.215%

    No Known Activations