© 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. 90990
    Prev
    Next
    INDEX
    Explanations

    resource-efficient

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

    The neuron fires on numeric tokens or numeric expressions (e.g. plain numbers, multi‐digit values, and hyphenated numeric ranges).

    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
    itisation
    -0.93
    ppling
    -0.93
    activating
    -0.87
    detection
    -0.86
     to
    -0.86
     potencial
    -0.85
    vább
    -0.85
    stuck
    -0.85
    üsseldorf
    -0.85
    capability
    -0.85
    POSITIVE LOGITS
     friendly
    1.61
     sensitive
    1.40
     efficient
    1.38
     agnostic
    1.36
     neutral
    1.29
     invariant
    1.27
     insensitive
    1.18
     resistant
    1.12
     tolerant
    1.11
    ově
    1.08
    Activations Density 0.287%

    No Known Activations