© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 11100
    Prev
    Next
    INDEX
    Explanations

    The neuron flags words and phrases typical of legal‐policy language about granting or restricting rights—e.g. “sell,” “permit,” “disclose,” “reserve” and related permission-and-restriction terms.

    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
    This
    0.60
     ovom
    0.60
     ovoj
    0.58
    avalt
    0.56
    hilangan
    0.52
     selle
    0.52
    திலும்
    0.52
    苚
    0.52
     这是
    0.51
    ̓
    0.51
    POSITIVE LOGITS
     such
    5.22
    such
    4.75
    Such
    4.50
     SUCH
    4.44
     Such
    4.41
    这样的
    4.00
     такие
    3.89
     таких
    3.88
     چنین
    3.78
     sådan
    3.69
    Activations Density 0.942%

    No Known Activations