© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76894
    Prev
    Next
    INDEX
    Explanations

    medical or technical queries

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

    technical file paths, system variables, and programming syntax in computing-related text.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron activates on uppercase technical identifiers—acronyms, file‐and‐include names, environment‐variable or code 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
    mickey
    -0.76
    Isolated
    -0.72
     コピー
    -0.71
    djang
    -0.71
    ropriate
    -0.70
    žití
    -0.69
     Taunton
    -0.69
    énégal
    -0.69
     Akron
    -0.68
     特徴
    -0.68
    POSITIVE LOGITS
    reply
    1.01
     reply
    0.91
    Reply
    0.87
     РС
    0.84
    beitung
    0.82
    ография
    0.81
    KeyPress
    0.78
     propOrder
    0.77
    valu
    0.77
     Flugzeug
    0.75
    Activations Density 0.007%

    No Known Activations