© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 68228
    Prev
    Next
    INDEX
    Explanations

    entity names and labels

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

    This neuron activates on code identifiers and property names—especially occurrences of “Name,” “Title,” and related accessor or transformer methods (e.g. getName, ToString, lower) 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
    ++);
    -1.01
    NSNumber
    -0.85
     лицом
    -0.84
    研
    -0.80
     elles
    -0.80
    рет
    -0.79
     Cafe
    -0.79
     behavior
    -0.78
    lässlich
    -0.78
     pleaf
    -0.78
    POSITIVE LOGITS
     name
    0.97
    entikan
    0.92
    États
    0.91
    hael
    0.90
    ウォーター
    0.90
    ctif
    0.86
    酝
    0.85
    淒
    0.85
    整個
    0.84
    jarati
    0.84
    Activations Density 0.031%

    No Known Activations