© 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. 22-GEMMASCOPE-RES-131K
    7. 79531
    Prev
    Next
    INDEX
    Explanations

    recursive programming concepts

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

    The neuron fires strongly on the substrings “recurs” (as in “Recursion,” “recursive,” etc.), i.e. it detects mentions of recursion.

    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
     одной
    -2.55
     technischen
    -2.27
     This
    -2.20
     Your
    -2.09
     einzelnen
    -2.08
    ᠌
    -2.08
    干什么
    -2.05
    翃
    -2.00
     dont
    -1.95
    干嘛
    -1.95
    POSITIVE LOGITS
    end
    2.66
    ”
    2.61
    ”،
    2.48
    that
    2.42
    get
    2.22
    ！”
    2.20
    縯
    2.13
    ”，
    2.08
    an
    2.06
    the
    2.06
    Activations Density 0.013%

    No Known Activations