© 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. 8981
    Prev
    Next
    INDEX
    Explanations

    options, vectors, transitions, actions

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

    This neuron primarily detects common nouns referring to items or resources (e.g., “options,” “website,” “file,” “types,” etc.).

    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
    顕
    0.58
    twin
    0.57
    uterine
    0.57
     FIF
    0.56
     হোয়াটসঅ্যাপ
    0.55
    बूझ
    0.55
    bandwidth
    0.54
    ൺ
    0.54
    ാൽ
    0.53
    ]'
    0.53
    POSITIVE LOGITS
    ازی
    0.70
     goof
    0.69
     svoj
    0.66
     swoim
    0.66
     végétaux
    0.66
     effectué
    0.66
     endroit
    0.65
     tuy
    0.64
     svoje
    0.64
    业主
    0.64
    Activations Density 0.000%

    No Known Activations