© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99632
    Prev
    Next
    INDEX
    Explanations

    right or straight direction

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

    The neuron is strongly triggered by the adverb “right.”

    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
    øse
    -0.87
    èle
    -0.83
    ΜΑ
    -0.82
    әр
    -0.80
    zko
    -0.79
    屐
    -0.77
     entirely
    -0.76
    coming
    -0.74
     Capri
    -0.73
     Connie
    -0.73
    POSITIVE LOGITS
     into
    1.20
    ставьте
    0.88
     revelou
    0.86
     up
    0.82
     其他
    0.82
    倒是
    0.81
     nasceu
    0.79
    KommentareTeilen
    0.78
     mostrou
    0.78
    nadequate
    0.77
    Activations Density 0.032%

    No Known Activations