© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 20-GEMMASCOPE-TRANSCODER-16K
    7. 10406
    Prev
    Next
    INDEX
    Explanations

    words and phrases connected to actions or states involving people, often emphasizing movement, observation, or influence.

    oai_token-act-pair · gemini-2.0-flash

    walking

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    strzy
    -0.47
    تمع
    -0.46
     szcz
    -0.45
     sucedido
    -0.44
     htons
    -0.44
    asca
    -0.43
    Slee
    -0.42
     Hiller
    -0.42
    setz
    -0.42
    amaño
    -0.42
    POSITIVE LOGITS
     walks
    1.20
     walking
    1.18
     walk
    1.16
     wandering
    1.15
     wander
    1.12
    walks
    1.09
     Walks
    1.07
    walk
    1.05
     Walking
    1.04
    walking
    1.03
    Activations Density 1.769%

    No Known Activations