© 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. 10-GEMMASCOPE-RES-131K
    7. 13116
    Prev
    Next
    INDEX
    Explanations

    deep learning and neural networks

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

    The main thing this neuron does is detect mentions of deep learning or neural network architecture terminology.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     would
    -1.62
    Despite
    -1.52
     one
    -1.46
    When
    -1.46
     situations
    -1.45
     knew
    -1.41
     despite
    -1.41
     situation
    -1.39
     although
    -1.38
     when
    -1.38
    POSITIVE LOGITS
    盉
    1.39
     inserir
    1.38
    
    1.26
     dando
    1.26
    нская
    1.24
    currently
    1.24
     atualizar
    1.23
    EDAD
    1.23
     verminder
    1.22
    ゲージ
    1.22
    Activations Density 0.046%

    No Known Activations