© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 1923
    Prev
    Next
    INDEX
    Explanations

    the word "letter" and related words about writing

    oai_token-act-pair · gemini-2.0-flash

    references to letters or written communication

    np_token-act-pair-logits · gpt-4o-mini

    letter

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    kuuta
    -0.61
    stateMutability
    -0.57
    mobileqq
    -0.56
    PhysRevLett
    -0.52
    Bibliographie
    -0.50
    DEAR
    -0.49
    
    -0.48
     Majefty
    -0.48
    alya
    -0.48
    gehalt
    -0.47
    POSITIVE LOGITS
    SequentialGroup
    0.81
    FormTagHelper
    0.64
    principalTable
    0.63
    RegistryLite
    0.60
    ///</
    0.59
    contentLoaded
    0.56
     useAppContext
    0.56
     transférez
    0.55
     незавершена
    0.55
    SharedCtor
    0.54
    Activations Density 3.014%

    No Known Activations