© 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. 31-GEMMASCOPE-2-RES-16K
    7. 12617
    Prev
    Next
    INDEX
    Explanations

    developers, researchers, designers

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

    This neuron fires on the formal legal‐style allowance phrase “at the discretion of,” i.e. boilerplate legal/disclaimer wording.

    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
     clientele
    0.60
     civilian
    0.56
     व्यक्ती
    0.55
     tenant
    0.55
     ক্ষমতায়
    0.55
    協力
    0.55
     பக்த
    0.54
    行业的
    0.54
    უშა
    0.53
     ব্যক্তির
    0.53
    POSITIVE LOGITS
     investigators
    1.88
     researchers
    1.86
     authors
    1.70
     designers
    1.68
    Investigators
    1.57
    Researchers
    1.55
     investigadores
    1.54
     analysts
    1.50
     Researchers
    1.49
     scientists
    1.44
    Activations Density 0.651%

    No Known Activations