© 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. 33015
    Prev
    Next
    INDEX
    Explanations

    William Blake poems/songs

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

    The neuron is detecting titles of literary works (especially well‐known poems) or their opening lines.

    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
    udin
    -0.77
     глу
    -0.72
    óloga
    -0.71
    Roche
    -0.71
    Marion
    -0.69
     Halliday
    -0.69
     barometer
    -0.68
     owed
    -0.68
     decided
    -0.68
     getAll
    -0.67
    POSITIVE LOGITS
     Blake
    1.07
    ↤
    0.96
    Blake
    0.93
     Songs
    0.93
    Songs
    0.92
     Tyre
    0.79
     Innocence
    0.79
     Sweden
    0.77
     Chimney
    0.76
     Albion
    0.75
    Activations Density 0.013%

    No Known Activations