© 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. 34-GEMMASCOPE-RES-131K
    7. 117566
    Prev
    Next
    INDEX
    Explanations

    Arabist, New Zealand, Che Guevara

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

    The neuron flags content-bearing (informative) words—i.e. non-stopwords such as names, dates, nouns, and verbs—while ignoring function words.

    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
    屹
    -0.90
    ileo
    -0.76
    籠
    -0.70
    Missile
    -0.69
     IgM
    -0.69
    FolderPath
    -0.69
    Clover
    -0.68
    losseum
    -0.68
    olympic
    -0.68
     quantitative
    -0.67
    POSITIVE LOGITS
     Lawrence
    1.34
    Lawrence
    1.34
     LAWRENCE
    1.20
     Revolt
    1.11
     desert
    1.08
     deserts
    1.01
    Arab
    1.01
     Arabia
    0.96
    WRENCE
    0.93
     Arab
    0.93
    Activations Density 0.015%

    No Known Activations