© 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. 792
    Prev
    Next
    INDEX
    Explanations

    make, split, raw, executable, empty

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

    The neuron fires on occurrences of “make” (in any case) in code contexts—e.g. in function or rule names like “MakeSplit,” “MakeRaw,” “make Wrapper,” or “make file ….”

    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
     minimally
    -0.79
     folate
    -0.76
    种子
    -0.73
    Fon
    -0.72
     upside
    -0.71
     Zwischen
    -0.71
    коменду
    -0.70
    DDT
    -0.69
     kleur
    -0.69
     rozwiąz
    -0.68
    POSITIVE LOGITS
    difference
    0.87
    orian
    0.84
     lasting
    0.79
    SHELL
    0.79
    tischen
    0.77
    made
    0.75
    BufferedReader
    0.75
     appearances
    0.74
     употреб
    0.74
    ISTA
    0.74
    Activations Density 0.048%

    No Known Activations