© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 80994
    Prev
    Next
    INDEX
    Explanations

    poetic meter and musical meter

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

    The neuron is reliably firing on mentions of literary genres and related descriptive labels (e.g. poetry, prose, tragedy) in historical or stylistic contexts.

    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
    babel
    -0.81
    loop
    -0.80
     ddelweddau
    -0.76
     loop
    -0.70
    ;$
    -0.69
    畢
    -0.68
     Boswell
    -0.68
    olithic
    -0.66
    ESD
    -0.66
     Loop
    -0.65
    POSITIVE LOGITS
     meter
    2.69
     metre
    2.44
     meters
    2.33
     Meter
    2.20
    meter
    2.13
    Meter
    2.11
     METER
    1.97
     metres
    1.95
    metre
    1.91
    meters
    1.82
    Activations Density 0.028%

    No Known Activations