© 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. 36126
    Prev
    Next
    INDEX
    Explanations

    code snippets and code-related text

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

    The neuron fires on code tokens that represent object method calls (i.e. a dot-prefixed identifier followed by parentheses).

    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
     Nip
    -0.91
    `,
    -0.85
    ヌー
    -0.84
    ^^^^^^^^
    -0.83
    ^^^^
    -0.82
    hysema
    -0.81
    咂
    -0.80
    貲
    -0.80
     with
    -0.79
     Excise
    -0.78
    POSITIVE LOGITS
     Overall
    0.96
     overall
    0.92
     cannot
    0.90
     it
    0.86
    Overall
    0.85
    chaqueta
    0.83
     makinesi
    0.82
     consoles
    0.80
    Sö
    0.79
     would
    0.78
    Activations Density 0.007%

    No Known Activations