© 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. 34-GEMMASCOPE-RES-131K
    7. 30546
    Prev
    Next
    INDEX
    Explanations

    faded stars and achievers

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

    This neuron strongly activates on numeric tokens (digits and numbers) and related list/table entries.

    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
    ṣ
    -0.82
     проч
    -0.80
     eyeballs
    -0.79
     infernal
    -0.77
    yttä
    -0.76
    ॰
    -0.75
    ÊNCIAS
    -0.75
     flera
    -0.75
     пароль
    -0.75
    farers
    -0.74
    POSITIVE LOGITS
     washed
    2.91
    washed
    2.09
     aging
    1.79
     fading
    1.79
     faded
    1.78
     has
    1.62
     ageing
    1.61
     failed
    1.55
     retired
    1.52
     wash
    1.48
    Activations Density 0.047%

    No Known Activations