© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32922
    Prev
    Next
    INDEX
    Explanations

    assume or play roles

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

    The neuron primarily activates on numeric tokens or figures (i.e., digits and counts) in the 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_34/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
     гу
    -0.74
    Endless
    -0.71
    ズム
    -0.70
    ECI
    -0.69
     celebró
    -0.66
    ุด
    -0.65
     dunk
    -0.65
     Hond
    -0.64
    に通
    -0.63
    足
    -0.63
    POSITIVE LOGITS
     assume
    2.39
     undertake
    2.19
    assume
    2.03
     assumed
    2.02
     assuming
    2.00
     undertaken
    2.00
     assumes
    1.89
    担
    1.84
     undertaking
    1.84
     assumption
    1.80
    Activations Density 0.033%

    No Known Activations