© 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. 10-GEMMASCOPE-RES-131K
    4. 111100
    Prev
    Next
    INDEX
    Explanations

    code snippets and questions

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

    the opening brace of an error-checking conditional block in code (specifically after "if err != nil").

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron detects error‐checking conditionals (e.g. “if err != nil { … }” or “if (error) …”) 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_10/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
     héroe
    -1.47
     grossen
    -1.28
    ⾼
    -1.20
    Возраст
    -1.20
    寕
    -1.19
    temporada
    -1.19
     gänzlich
    -1.18
    bitch
    -1.17
     Continued
    -1.16
     numerosas
    -1.16
    POSITIVE LOGITS
     and
    1.68
     they
    1.29
    ्टी
    1.24
     outros
    1.24
     otros
    1.23
    چار
    1.23
    tiga
    1.22
    лакти
    1.21
    Antworten
    1.16
     corolla
    1.16
    Activations Density 0.004%

    No Known Activations