© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27452
    Prev
    Next
    INDEX
    Explanations

    code or specific introductory phrases

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

    The neuron is looking for code syntax elements—particularly function/method invocations and conditional checks (e.g. “if(…)", bracketed calls, and method 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
    Takk
    -1.10
    Люди
    -1.08
     tipis
    -1.03
    rätt
    -1.03
    quels
    -0.98
    Ordre
    -0.98
     唯美
    -0.94
     čer
    -0.94
     Launched
    -0.93
    给他们
    -0.93
    POSITIVE LOGITS
     this
    1.29
    IllegalArgument
    1.13
     choć
    1.12
     acceptable
    1.06
     can
    1.05
     S
    1.02
     if
    1.01
     like
    0.99
    <em>
    0.96
     D
    0.95
    Activations Density 0.004%

    No Known Activations