© 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. 12570
    Prev
    Next
    INDEX
    Explanations

    beginnings of code or structured text

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

    The neuron is tuned to detect occurrences of the “if” keyword (i.e. conditional statements) 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
    でしたか
    -0.93
     adquirido
    -0.92
     चीज़ों
    -0.92
    בודה
    -0.91
    させていただきました
    -0.91
     my
    -0.91
     kriminal
    -0.90
     atacado
    -0.88
    altro
    -0.85
    staus
    -0.85
    POSITIVE LOGITS
     doesn
    1.02
    襌
    0.97
     isn
    0.93
     evitare
    0.93
    remaining
    0.93
     données
    0.91
     conservación
    0.91
     помощ
    0.91
     sugerencias
    0.90
     allarg
    0.90
    Activations Density 0.000%

    No Known Activations