© 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. 10-GEMMASCOPE-RES-131K
    4. 105364
    Prev
    Next
    INDEX
    Explanations

    making promises

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

    The neuron strongly activates on words expressing a pledge or promise (e.g. “promise,” “promises,” “pledged,” “promised”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    hamilan
    -1.76
     TAMBÉM
    -1.57
     pól
    -1.49
     pú
    -1.45
    conocimiento
    -1.45
    adillas
    -1.45
     drodze
    -1.41
    昨日
    -1.41
    cuerdo
    -1.41
     glä
    -1.40
    POSITIVE LOGITS
    or
    1.85
    as
    1.84
    et
    1.73
     he
    1.72
    is
    1.60
     However
    1.57
    al
    1.55
    ine
    1.53
    ot
    1.52
    not
    1.51
    Activations Density 0.008%

    No Known Activations