© 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. 33215
    Prev
    Next
    INDEX
    Explanations

    code snippets or comments

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

    The neuron responds strongly to structural tokens—especially angle-bracketed tags (e.g. “<p>” or “</p>”) and standalone numeric tokens.

    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
    熵
    -0.75
     Scal
    -0.73
     strlen
    -0.69
    Zz
    -0.69
    addington
    -0.67
    レート
    -0.67
     Priya
    -0.66
    lý
    -0.65
     refer
    -0.65
     Scheme
    -0.65
    POSITIVE LOGITS
     canciones
    0.69
    raseña
    0.68
     Magee
    0.68
     caen
    0.64
     गु
    0.63
    ୨
    0.63
     penonton
    0.63
    Eigenschaften
    0.63
    Telefono
    0.62
    గు
    0.61
    Activations Density 0.069%

    No Known Activations