© 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. 22-GEMMASCOPE-RES-131K
    4. 12464
    Prev
    Next
    INDEX
    Explanations

    variable names and indices

    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
     from
    -1.39
     with
    -1.29
     or
    -1.23
     while
    -1.15
     like
    -1.13
     by
    -1.09
     getting
    -1.02
     With
    -0.99
     at
    -0.97
     Unlike
    -0.97
    POSITIVE LOGITS
    immediately
    1.12
     ſou
    1.10
    ulence
    1.10
    そういえば
    1.08
     agres
    1.06
     equivalente
    1.05
    
    1.05
    ちなみに
    1.05
    Mvh
    1.04
     ALONE
    1.04
    Activations Density 0.002%

    No Known Activations