© 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. 34-GEMMASCOPE-RES-131K
    7. 99735
    Prev
    Next
    INDEX
    Explanations

    author instructions

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

    The neuron spikes on structural formatting markers in academic texts—section headers, separators (lines of “=” or “*”), numbered labels, and similar layout tokens.

    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
     didn
    -0.80
     poveznice
    -0.79
    BIBLIOGRAPHY
    -0.76
    món
    -0.76
     today
    -0.75
    programmer
    -0.75
     bibliography
    -0.75
     Implementation
    -0.75
     produced
    -0.74
    Filmography
    -0.73
    POSITIVE LOGITS
     manuscript
    1.62
     authors
    1.35
     manuscripts
    1.34
     submission
    1.28
    manus
    1.24
    authors
    1.21
    Authors
    1.14
     submissions
    1.13
     submitting
    1.11
     manuscrit
    1.10
    Activations Density 0.062%

    No Known Activations