© 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. 28460
    Prev
    Next
    INDEX
    Explanations

    string conversion

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

    The neuron fires on references to type‐conversion routines—especially string‐to‐number and number‐to‐string conversion calls.

    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
     tā
    -1.00
    聼
    -0.99
    を行う
    -0.93
     understand
    -0.92
     maría
    -0.91
    icken
    -0.91
    级
    -0.88
    鹄
    -0.88
    とな
    -0.88
    isty
    -0.87
    POSITIVE LOGITS
     of
    1.44
     aloud
    1.09
    imetsu
    1.00
    интере
    0.94
    occuper
    0.92
    0.92
    erobe
    0.91
     Resultat
    0.91
    山頂
    0.90
     concedes
    0.90
    Activations Density 0.004%

    No Known Activations