© 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. 34-GEMMASCOPE-RES-131K
    7. 97736
    Prev
    Next
    INDEX
    Explanations

    Objective-C `@implementation` blocks

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

    This neuron fires on runs of whitespace (indentation and spacing), especially the leading spaces/tabs before code block delimiters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     revisor
    -0.87
    marginHorizontal
    -0.87
    ariConfig
    -0.85
    捗
    -0.82
     until
    -0.80
     Betten
    -0.79
    🙂
    -0.77
     whose
    -0.77
     Hallen
    -0.76
    UIButton
    -0.76
    POSITIVE LOGITS
     Marsha
    0.87
     Gewähr
    0.82
     nötig
    0.82
    当然是
    0.80
     récemment
    0.78
    mės
    0.77
    ヴェル
    0.77
     Elsie
    0.77
     сохра
    0.76
     brasileiro
    0.76
    Activations Density 0.004%

    No Known Activations