© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71157
    Prev
    Next
    INDEX
    Explanations

    neural network components and context identifiers

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

    The neuron strongly activates on mentions of “encoder” (especially in “autoencoder”) and related model components like “decoder.”

    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
     oogen
    -1.47
     they
    -1.31
     lorenzo
    -1.30
     手順
    -1.30
    麃
    -1.29
     imbal
    -1.27
     lada
    -1.27
     alberi
    -1.27
    正確
    -1.26
     Nc
    -1.26
    POSITIVE LOGITS
    ")
    1.58
     cbd
    1.51
    ()],
    1.50
    ()]);
    1.48
     todas
    1.45
    //}
    
    1.45
    }
    
    1.44
    ſelves
    1.42
    ()])
    1.39
     singles
    1.34
    Activations Density 0.009%

    No Known Activations