© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25467
    Prev
    Next
    INDEX
    Explanations

    coding syntax and structure

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

    The neuron is primarily triggered by numeric literal tokens (i.e. numbers) in code.

    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
     خریدار
    -1.26
     tatuaje
    -1.23
    時間で
    -1.20
    ּוֹ
    -1.14
     cordón
    -1.13
     februari
    -1.10
     данного
    -1.10
    りたい
    -1.09
    ִּ
    -1.08
     received
    -1.06
    POSITIVE LOGITS
    смарт
    1.23
     лимон
    1.22
    ⎧
    1.16
     territ
    1.14
    aneity
    1.13
     aiutare
    1.13
     entweder
    1.11
    无论是
    1.08
    大多
    1.07
    jacking
    1.07
    Activations Density 0.004%

    No Known Activations