© 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. 25566
    Prev
    Next
    INDEX
    Explanations

    intellij, github, trademarks, Note

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

    This neuron especially lights up on tokens that are names or domains of software projects or organizations (e.g. “apache,” “intellij,” “githubusercontent,” etc.).

    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
     cero
    -0.84
    vish
    -0.76
    zero
    -0.75
     zero
    -0.74
    finish
    -0.66
     Finish
    -0.66
     achet
    -0.65
    ෂ
    -0.65
    prud
    -0.65
    Zero
    -0.65
    POSITIVE LOGITS
     Glove
    0.69
    copio
    0.66
    ToUser
    0.65
     THEOREM
    0.65
     Investigate
    0.65
    itiers
    0.64
     Geothermal
    0.64
     Lowest
    0.63
     Passenger
    0.63
     இ
    0.62
    Activations Density 0.079%

    No Known Activations