© 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. 22-GEMMASCOPE-RES-131K
    7. 72027
    Prev
    Next
    INDEX
    Explanations

    separate steps

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

    This neuron activates on words that signal dividing content into parts or procedural steps (e.g. divided, into, development, separate).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    分に
    -0.90
     Kazakhstan
    -0.90
     They
    -0.88
    мл
    -0.87
     Temporary
    -0.86
     tope
    -0.80
     Today
    -0.79
     testa
    -0.79
     Kenya
    -0.79
    Significado
    -0.78
    POSITIVE LOGITS
     time
    1.29
     procedure
    1.20
     strategy
    1.10
     approach
    1.09
     happening
    1.07
     discussions
    1.06
     steps
    1.06
     discussion
    1.05
     doing
    1.01
     issue
    1.00
    Activations Density 0.199%

    No Known Activations