© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 8594
    Prev
    Next
    INDEX
    Explanations

    a mix of words related to: 1) dates and times, 2) locations/geopolitics, 3) document metadata/references and 4) people's professions.

    oai_token-act-pair · gemini-2.0-flash

    Countries and cultures

    np_max-act-logits · gemini-2.0-flash

    say "nationalities"

    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_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
     viewDidLoad
    -0.68
     IndexPath
    -0.63
    脚注の使い方
    -0.63
    MethodManager
    -0.60
     Korean
    -0.60
    ArrowToggle
    -0.60
    XmlAccessorType
    -0.59
    WillAppear
    -0.59
    ConstraintMaker
    -0.57
    GeoNames
    -0.56
    POSITIVE LOGITS
     Cong
    1.16
     Cam
    1.08
     Gamb
    1.04
     Mal
    1.03
     Liber
    1.02
     Bos
    0.91
     Ang
    0.89
     Maur
    0.88
     Mold
    0.86
     Tog
    0.85
    Activations Density 6.020%

    No Known Activations