© 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. 10-GEMMASCOPE-RES-131K
    7. 105111
    Prev
    Next
    INDEX
    Explanations

    stag do, postmortem, stimulating

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

    The neuron is highly sensitive to tokens that serve as labels immediately before quantitative values—that is, words used as headings or descriptors directly followed by numbers (e.g. “Stag,” “postmortem,” “stim”) signaling measurement data.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    1
    -3.25
    A
    -2.94
    5
    -2.86
    8
    -2.81
    7
    -2.58
    鳚
    -2.52
    ずっと
    -2.44
    0
    -2.33
    6
    -2.33
    4
    -2.30
    POSITIVE LOGITS
     is
    2.73
    ISHMENT
    2.53
     particular
    2.52
     crees
    2.50
    y
    2.42
    嗞
    2.42
     próximos
    2.38
    在住
    2.36
     рус
    2.34
     cómod
    2.31
    Activations Density 0.010%

    No Known Activations