© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 11284
    Prev
    Next
    INDEX
    Explanations

    ends in 'of', 'on', 'to', 'with', 'down'

    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_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
     their
    -1.41
     any
    -1.24
     five
    -1.23
     six
    -1.23
     each
    -1.23
    Fonto
    -1.21
     if
    -1.21
     use
    -1.20
     this
    -1.20
     new
    -1.18
    POSITIVE LOGITS
     همچنین
    1.28
     allerlei
    1.19
     بشكل
    1.16
     identif
    1.11
     legendar
    1.10
    panes
    1.08
     alcune
    1.06
    觊
    1.05
    ·
    1.05
     SOMETHING
    1.05
    Activations Density 0.058%

    No Known Activations