© 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. 11316
    Prev
    Next
    INDEX
    Explanations

    a variety of punctuation marks and conjunctions that separate pieces of information or indicate relationships between them.

    oai_token-act-pair · gemini-2.0-flash

    conjunctions or introductory phrases

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

    punctuation

    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
    ,
    -0.77
    +#+#
    -0.55
     disponibilités
    -0.50
    ROIT
    -0.46
    、
    -0.46
    ，
    -0.44
    Hej
    -0.44
    役
    -0.43
    令
    -0.43
     تضيفلها
    -0.42
    POSITIVE LOGITS
     which
    1.13
     a
    0.94
     part
    0.81
     onderdeel
    0.75
     an
    0.74
    "])
    0.73
     whoſe
    0.73
     another
    0.71
     WHICH
    0.71
    which
    0.70
    Activations Density 4.564%

    No Known Activations