© 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. 31-GEMMASCOPE-2-RES-16K
    7. 11362
    Prev
    Next
    INDEX
    Explanations

    bullet points or list

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

    The neuron detects mentions of geographical gulfs, i.e. the phrase “Gulf of” followed by a place name.

    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_31_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
    <blockquote>
    0.63
     hey
    0.57
     incididunt
    0.54
    što
    0.52
    ][$
    0.52
    ][]
    0.52
    ழுத
    0.52
     WoW
    0.52
     Wow
    0.51
    失
    0.51
    POSITIVE LOGITS
    Spoiler
    0.70
    狀態
    0.63
     ​
    0.57
     opio
    0.57
     oto
    0.57
     anbieten
    0.57
    .​
    0.56
    獸
    0.56
    號
    0.54
    账号
    0.53
    Activations Density 0.022%

    No Known Activations