© 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-2-2B
    5. 2-GEMMASCOPE-TRANSCODER-16K
    7. 13586
    Prev
    Next
    INDEX
    Explanations

    words ending in 'thing' and 'one' often with preceding words indicating the presence of something

    oai_token-act-pair · gemini-2.0-flash

    thin

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_2/width_16k/average_l0_49
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.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
    бливості
    -0.59
    armani
    -0.57
    Identyfik
    -0.56
    WebElementEntity
    -0.55
    SPATH
    -0.54
    wifery
    -0.53
    anio
    -0.53
    Affect
    -0.52
    Datuak
    -0.52
    jgl
    -0.50
    POSITIVE LOGITS
    <bos>
    1.44
     anyone
    1.14
     everyone
    1.14
     everybody
    1.09
    Everyone
    1.07
    Everybody
    1.06
    everyone
    1.05
     anybody
    1.04
     Anyone
    1.03
     Everybody
    1.02
    Activations Density 0.411%

    No Known Activations