© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 103579
    Prev
    Next
    INDEX
    Explanations

    self-consistent approach or treatment

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

    This neuron strongly activates on specialized, technical physics method and analysis terms (e.g., “mean‐field,” “self‐consistent,” “variational,” “renormalization,” “field theory”).

    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
     Merc
    -0.84
    ⍴
    -0.81
     legislators
    -0.79
    ‽
    -0.78
     downloads
    -0.77
     GLP
    -0.77
     الحكومة
    -0.77
     Haem
    -0.76
     tast
    -0.75
     Legisl
    -0.75
    POSITIVE LOGITS
     treatment
    1.41
     Treatment
    1.23
     treating
    1.13
    Treatment
    1.11
     treat
    1.10
    treatment
    1.09
     treatments
    1.09
     approach
    1.09
     approximation
    1.04
    treat
    1.02
    Activations Density 0.127%

    No Known Activations