© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 1668
    Prev
    Next
    INDEX
    Explanations

    words ending in "ge/ges", verbs ending in "ged/ging", and the words "dot" , "=" , "
" and "log" (possibly due to code)

    oai_token-act-pair · gemini-2.0-flash

    indul

    np_max-act-logits · gemini-2.0-flash

    say "g" words Method used: 2 — because all following pieces begin with g; say "g" words

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     autorytatywna
    -0.58
    Хьажоргаш
    -0.54
     незавершена
    -0.50
    queryInterface
    -0.50
    contentLoaded
    -0.49
    windowFixed
    -0.48
    WriteAttribute
    -0.48
    IVEREF
    -0.47
    AutoScaleMode
    -0.46
     colspan
    -0.46
    POSITIVE LOGITS
    gw
    0.92
    gs
    0.90
    g
    0.85
    gu
    0.84
    gue
    0.84
    gv
    0.83
    gt
    0.83
    gas
    0.82
    ged
    0.82
    ga
    0.82
    Activations Density 5.516%

    No Known Activations