© 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. 3836
    Prev
    Next
    INDEX
    Explanations

    complex mathematical jargon and technical terms

    oai_token-act-pair · gemini-2.0-flash

    code and math

    np_max-act-logits · gemini-2.0-flash

    say "punctuation"

    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
     ,
    -2.58
    ，
    -1.70
     ，
    -1.43
     ,
    -1.43
     ،
    -1.35
     ,\
    -1.26
    .,
    -1.17
    (),
    -1.11
    ™,
    -1.09
     ,"
    -1.09
    POSITIVE LOGITS
    abetes
    0.59
     itſelf
    0.52
     Eſ
    0.51
    .');
    0.50
    хь
    0.48
     ʻ
    0.48
     §.
    0.48
    zysta
    0.47
    hô
    0.47
    InstanceState
    0.47
    Activations Density 5.174%

    No Known Activations