© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 60044
    Prev
    Next
    INDEX
    Explanations

    open/close operations

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

    The neuron responds strongly to code tokens for low‐level file or resource operations—especially the calls “open” and “close.”

    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
     to
    -0.98
    しも
    -0.73
    Letting
    -0.70
    chito
    -0.70
     終
    -0.69
    なきゃ
    -0.67
    終
    -0.67
    cokinetics
    -0.66
    lfloor
    -0.66
    ποι
    -0.66
    POSITIVE LOGITS
    open
    1.02
    bewah
    0.93
     OPEN
    0.89
    TryDecode
    0.86
     aprire
    0.85
    앓
    0.83
     itd
    0.82
    塍
    0.82
    holdet
    0.82
     Leve
    0.80
    Activations Density 0.026%

    No Known Activations