© 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. 12233
    Prev
    Next
    INDEX
    Explanations

    band names and members

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

    This neuron fires on densely structured “infobox”-style lines—tokens from proper names, titles, dates, and numeric fields in the document’s metadata.

    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
     biar
    -0.85
    olut
    -0.73
    ROOM
    -0.73
    ∽
    -0.71
     cerebral
    -0.68
    pher
    -0.68
    grads
    -0.67
    nání
    -0.67
     Plato
    -0.67
     Philanth
    -0.67
    POSITIVE LOGITS
     Foo
    1.69
    Foo
    1.58
    FOO
    1.19
     Hawkins
    1.05
     foo
    1.02
     Gro
    1.00
    foo
    0.96
     Dave
    0.81
     Fighters
    0.79
    Gro
    0.74
    Activations Density 0.031%

    No Known Activations