© 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. 47712
    Prev
    Next
    INDEX
    Explanations

    host and CPU

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

    The neuron flags mentions of copying or transferring data back and forth between Host and Device (e.g. “from … to …” in CUDA memory transfers).

    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
    甙
    -0.73
    pkey
    -0.69
     ligiloj
    -0.68
     ව
    -0.68
    нке
    -0.68
     świąte
    -0.68
    เสียง
    -0.67
    وزه
    -0.66
    íaca
    -0.65
    Rhode
    -0.65
    POSITIVE LOGITS
     CPU
    2.92
    cpu
    2.78
     cpu
    2.77
    CPU
    2.61
     host
    2.59
    Host
    2.27
     Host
    2.27
    host
    2.14
    Cpu
    2.05
    HOST
    1.86
    Activations Density 0.037%

    No Known Activations