© 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. 10-GEMMASCOPE-RES-131K
    7. 71800
    Prev
    Next
    INDEX
    Explanations

    next steps and subsequent actions

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

    The neuron flags occurrences of “next step” or “next steps” phrases in procedural or instructional 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
     целью
    -1.80
     the
    -1.77
     Sehr
    -1.70
    )」
    -1.66
     Nicht
    -1.64
    艹
    -1.63
     Niemand
    -1.49
    </th>
    -1.49
     yapmak
    -1.42
     sluit
    -1.41
    POSITIVE LOGITS
    ☫
    1.72
    ᜀ
    1.59
    an
    1.58
    他
    1.55
    𝑈
    1.50
     بشكل
    1.46
    movq
    1.45
    ჻
    1.45
     设备
    1.44
    ки
    1.42
    Activations Density 0.011%

    No Known Activations