© 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. 34-GEMMASCOPE-RES-131K
    4. 66388
    Prev
    Next
    INDEX
    Explanations

    programming and legal contexts

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

    The neuron strongly activates on words and punctuation typical of software‐license or permission‐grant boilerplate (e.g. “Permission,” “distribution,” “modification,” “permitted”).

    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
     Farina
    -0.68
     numeric
    -0.67
    了两
    -0.66
     لهم
    -0.66
     wereld
    -0.66
    上映
    -0.65
     Lime
    -0.65
     τε
    -0.63
     nơi
    -0.63
    pageX
    -0.63
    POSITIVE LOGITS
     zr
    0.74
     检查
    0.71
     Camilla
    0.69
     vince
    0.69
     poft
    0.67
     dosa
    0.67
    philly
    0.66
    Dos
    0.66
    মা
    0.65
     borse
    0.65
    Activations Density 0.054%

    No Known Activations