© 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. 23883
    Prev
    Next
    INDEX
    Explanations

    lect, epitaph, dialectic, architect, archetype

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

    The neuron spikes on the early character sequences of longer, multi‐syllable words—i.e. it detects long‐word prefixes.

    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
    창
    -0.74
     carol
    -0.70
    喰
    -0.70
    ╙
    -0.69
    styleUrls
    -0.68
    歿
    -0.68
     singled
    -0.66
     الجام
    -0.66
    ुप
    -0.66
    فيه
    -0.66
    POSITIVE LOGITS
     lect
    0.97
    LECT
    0.88
    ects
    0.84
    lecture
    0.80
    lect
    0.77
    ecture
    0.76
    URAL
    0.73
     lecture
    0.73
     tect
    0.71
    Theater
    0.70
    Activations Density 0.043%

    No Known Activations