© 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. 10229
    Prev
    Next
    INDEX
    Explanations

    on- words

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

    The neuron responds strongly to capitalized personal or place names—i.e. proper‐noun tokens.

    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
    スペシャル
    0.54
     मध्या
    0.54
    weapon
    0.53
     BETTER
    0.53
    イヤ
    0.53
     हथियार
    0.52
    isolate
    0.52
    밑
    0.52
    tell
    0.52
    candy
    0.52
    POSITIVE LOGITS
     On
    0.79
    slaught
    0.71
    갖
    0.69
     hiatus
    0.69
     Off
    0.65
    omat
    0.64
     Ons
    0.64
    ions
    0.63
     ON
    0.62
     isOn
    0.62
    Activations Density 0.145%

    No Known Activations