Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 14374
    Prev
    Next
    INDEX
    Explanations

    escaped backslashes possibly followed by the word "peer"

    oai_token-act-pair · gemini-2.0-flash

    Code/technical document snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     iſt
    -0.93
     pleaſure
    -0.93
     myſelf
    -0.90
     Monfieur
    -0.88
     raiſ
    -0.88
     Theſe
    -0.87
     Efq
    -0.86
     purpoſe
    -0.86
     houſe
    -0.85
     ſever
    -0.84
    POSITIVE LOGITS
    section
    0.54
    begin
    0.51
    <bos>
    0.50
     ISNI
    0.47
    %.
    0.47
    ora
    0.47
    $.
    0.46
    $
    0.46
    ri
    0.46
    ul
    0.45
    Activations Density 1.429%

    No Known Activations