© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 118524
    Prev
    Next
    INDEX
    Explanations

    programming keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     instead
    -1.13
     into
    -0.89
     ještě
    -0.88
     вместо
    -0.87
    tuy
    -0.85
    ())
    -0.83
    instead
    -0.82
    Dearest
    -0.81
     kvä
    -0.81
     säs
    -0.80
    POSITIVE LOGITS
     altogether
    0.97
    掉
    0.94
     unnecessary
    0.93
     any
    0.89
    이크
    0.89
     orice
    0.87
    rások
    0.86
     entirely
    0.85
    ГЛА
    0.84
     know
    0.82
    Activations Density 0.006%

    No Known Activations