© 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. 22-GEMMASCOPE-RES-131K
    7. 35757
    Prev
    Next
    INDEX
    Explanations

    comparison and contrast

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

    The neuron fires on longer, multi-syllable content tokens (e.g. technical terms, proper nouns, and other unusually long words).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     geht
    -0.77
    ل
    -0.74
    貞
    -0.73
    ̶
    -0.71
    </tbody>
    -0.70
    Growing
    -0.69
    迪
    -0.69
    最早
    -0.69
    Originally
    -0.69
    Must
    -0.69
    POSITIVE LOGITS
     inoltre
    1.04
     also
    0.98
     furthermore
    0.95
    izarea
    0.93
     secondly
    0.90
    araus
    0.89
    (&(
    0.89
     preuß
    0.88
     moreover
    0.88
    Ita
    0.86
    Activations Density 0.108%

    No Known Activations