© 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. 62883
    Prev
    Next
    INDEX
    Explanations

    cooling systems and apparatuses

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

    The neuron strongly activates on the word “cooling” and its close variants (e.g. “coolant”), i.e. technical mentions of cooling.

    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
    悄
    -0.90
    教堂
    -0.88
    воль
    -0.86
    omst
    -0.86
     للإ
    -0.84
    じゃあ
    -0.83
     отвечает
    -0.83
    uego
    -0.83
     キー
    -0.82
     crappy
    -0.82
    POSITIVE LOGITS
    cooling
    1.55
    Cooling
    1.34
     cooling
    1.23
     Cooling
    1.16
    FAN
    0.92
    ファン
    0.91
     down
    0.87
     system
    0.85
    ierend
    0.84
    .
    0.84
    Activations Density 0.015%

    No Known Activations