© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 13686
    Prev
    Next
    INDEX
    Explanations

    purpose or use

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

    This neuron fires on words that specify allowed-use restrictions and permissions (e.g. “personal,” “non-commercial,” “commercial,” “any,” “purpose”) in license or terms-of-use text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    -1.17
    thiet
    -1.13
     Şubat
    -1.13
    vaila
    -1.13
     Национа
    -1.09
    pakah
    -1.07
    惋
    -1.06
    legung
    -1.05
     of
    -1.04
    lessed
    -1.04
    POSITIVE LOGITS
     purposes
    2.61
     tasks
    1.86
     purpose
    1.77
     Purposes
    1.65
     functions
    1.53
     activities
    1.47
    purposes
    1.40
    用途
    1.36
     amaç
    1.34
     uses
    1.30
    Activations Density 0.050%

    No Known Activations