© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102884
    Prev
    Next
    INDEX
    Explanations

    Meta, #, Q, or Metacognition

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

    The neuron strongly activates on standalone numeric tokens (e.g. counts, statistics, line‐number values, decimal numbers) rather than regular words.

    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
    -3.72
    2
    -3.05
    3
    -2.91
    s
    -2.86
    -
    -2.69
    7
    -2.69
    4
    -2.59
     I
    -2.59
    6
    -2.55
    9
    -2.47
    POSITIVE LOGITS
     Pentru
    2.48
    🌸
    2.39
     Organizador
    2.28
    🤗
    2.25
     Autó
    2.23
     uitvo
    2.22
    ❤️❤️
    2.22
     renovar
    2.20
     berikan
    2.19
     体験
    2.17
    Activations Density 0.004%

    No Known Activations