© 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-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 10819
    Prev
    Next
    INDEX
    Explanations

    casual, informal, or relaxed tone

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

    The neuron fires strongly on exhortatory or motivational language—commands, warnings, or idiomatic calls to action.

    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
     जबर
    0.64
    czka
    0.61
     Forced
    0.57
     مجبور
    0.55
    ীব্র
    0.55
    跪
    0.54
    玾
    0.54
    捧
    0.53
     непоня
    0.52
     unanimously
    0.52
    POSITIVE LOGITS
     relaxed
    2.34
     relax
    2.27
     Relax
    2.16
    Relax
    2.14
     relaxing
    2.06
    relaxed
    2.06
     relaj
    2.02
    relax
    2.00
     relaxation
    1.97
     lazy
    1.97
    Activations Density 0.619%

    No Known Activations