© 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. 26629
    Prev
    Next
    INDEX
    Explanations

    length or size

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

    The neuron fires on code expressions that query or compare the size/length/count of a collection (e.g. calls to .length or len(), references to $# or PHP_INT_SIZE, and the numeric literals involved).

    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
     būs
    -1.11
    envolvimento
    -1.02
     with
    -1.00
     发型
    -0.95
     their
    -0.94
    Smit
    -0.90
     berpeng
    -0.90
    غيير
    -0.89
    ឌ
    -0.88
    Bret
    -0.88
    POSITIVE LOGITS
     is
    1.18
    トーン
    0.93
    起身
    0.89
    enfone
    0.89
     siquiera
    0.89
    ef
    0.88
     будущего
    0.86
    声が
    0.86
    𝑣
    0.84
    SC
    0.84
    Activations Density 0.002%

    No Known Activations