© 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. 90920
    Prev
    Next
    INDEX
    Explanations

    neighboring positions

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

    The neuron consistently activates on numeric tokens and technical specification terms (e.g. measurements, counts, radii, cell counts, etc.).

    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
    merksamkeit
    -0.82
    Интере
    -0.81
     solem
    -0.80
    Arrive
    -0.80
    Elton
    -0.77
    actéristiques
    -0.77
    mité
    -0.76
    醐
    -0.75
    oluto
    -0.75
    Volunteers
    -0.74
    POSITIVE LOGITS
     adjacent
    1.69
     surrounding
    1.63
     neighbors
    1.49
     neighbours
    1.43
     neighbor
    1.36
     neighborhood
    1.33
     neighbour
    1.27
     neighbourhood
    1.21
     neighboring
    1.20
     surround
    1.18
    Activations Density 0.020%

    No Known Activations