© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 6299
    Prev
    Next
    INDEX
    Explanations

    words appearing in a historical play script

    oai_token-act-pair · gemini-2.0-flash

    textual references related to books or titles

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

    Single characters

    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
     itſelf
    -1.55
     raiſ
    -1.39
     Theſe
    -1.35
     myſelf
    -1.34
    ſelves
    -1.31
    ſelf
    -1.30
     ſever
    -1.29
     themſelves
    -1.28
     auffi
    -1.26
     ſtate
    -1.26
    POSITIVE LOGITS
     t
    0.69
     
    0.67
    ,
    0.64
     (
    0.63
     da
    0.61
     G
    0.60
     g
    0.59
     T
    0.57
     as
    0.57
    <eos>
    0.56
    Activations Density 3.664%

    No Known Activations