© 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. 103762
    Prev
    Next
    INDEX
    Explanations

    github urls and repositories

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

    This neuron responds to structural layout cues—especially leading whitespace indentations and punctuation (colons, line prefixes) as used in YAML or other indented markup.

    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
     intitulé
    -0.73
     neuropathy
    -0.70
     infographics
    -0.69
    AXB
    -0.68
    rimiento
    -0.68
     casserole
    -0.66
    kish
    -0.66
    uropa
    -0.65
     Redirect
    -0.65
    ront
    -0.65
    POSITIVE LOGITS
     GitHub
    3.44
    GitHub
    2.92
     Github
    2.84
     github
    2.83
    github
    2.83
    Github
    2.47
     git
    2.13
     Git
    1.93
    GITHUB
    1.80
    git
    1.77
    Activations Density 0.144%

    No Known Activations