© 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. 115987
    Prev
    Next
    INDEX
    Explanations

    multiple layers

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

    The neuron strongly activates on any form of the word “layer” (e.g. layered, layering, multi-layered, layers).

    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
    nowned
    -0.98
    étape
    -0.93
     speckled
    -0.87
    Except
    -0.85
    ėmis
    -0.84
    로운
    -0.84
     sprigs
    -0.84
    VIATIONS
    -0.82
    skyl
    -0.81
    佔
    -0.80
    POSITIVE LOGITS
     layers
    1.77
     layer
    1.23
    layer
    1.19
     Layers
    1.18
    layers
    1.18
    Layer
    1.09
     thickness
    1.08
     Schichten
    1.08
    Layers
    1.08
     layered
    1.07
    Activations Density 0.026%

    No Known Activations