© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 40372
    Prev
    Next
    INDEX
    Explanations

    registry, yield, client, Random, Team, make

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

    The neuron detects numeric literals (multi‐digit numbers) in code, especially as assignment values.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    mua
    -1.02
     lã
    -1.02
     theſe
    -1.02
    looked
    -1.01
     toils
    -1.00
    hike
    -1.00
     nuvem
    -0.99
    lost
    -0.98
    oved
    -0.98
    iked
    -0.97
    POSITIVE LOGITS
     by
    1.44
     when
    1.23
     during
    1.10
     one
    1.09
    when
    1.07
     to
    1.06
     new
    1.05
    If
    1.05
     exuberant
    1.05
     in
    1.02
    Activations Density 0.003%

    No Known Activations