© 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. 43155
    Prev
    Next
    INDEX
    Explanations

    optimization objectives and likelihoods

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

    This neuron specializes in spotting mentions of mathematical objectives or loss‐function terms (e.g. “loss,” “function,” “objective,” “derivatives,” etc.) in technical or formulaic contexts.

    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
     who
    -2.02
    ógicos
    -1.67
    Privatpatient
    -1.53
     menganggap
    -1.52
    llavero
    -1.48
    :),
    -1.46
    лтемелер
    -1.45
    ®,
    -1.42
    әрмәләр
    -1.41
    gencias
    -1.41
    POSITIVE LOGITS
     it
    1.64
     two
    1.48
         
    1.48
     by
    1.44
     never
    1.38
     NOT
    1.38
     on
    1.35
    ↵↵↵
    1.34
     now
    1.30
      
    1.27
    Activations Density 0.018%

    No Known Activations