© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112368
    Prev
    Next
    INDEX
    Explanations

    the phrase is or has been

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

    The neuron responds to abstract nouns that mark key points or objectives in discourse (e.g. challenge, need, aim, intention, problem, emphasis, issue).

    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
    趼
    -1.68
     艺术
    -1.62
     the
    -1.59
    💧
    -1.50
     existant
    -1.50
    🔹
    -1.48
    Однако
    -1.48
    ورش
    -1.47
    🚗
    -1.44
    ☀️
    -1.42
    POSITIVE LOGITS
     about
    2.08
     to
    1.88
     against
    1.81
     innych
    1.73
     
    1.72
    0
    1.72
     predominantly
    1.71
     encontrada
    1.69
     інших
    1.68
     other
    1.66
    Activations Density 0.046%

    No Known Activations