© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 4761
    Prev
    Next
    INDEX
    Explanations

    a mix of sentence endings, section headers, capitalized words and some words associated with political sentiment or technical instructions

    oai_token-act-pair · gemini-2.0-flash

    By words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     Which
    -1.20
     Who
    -1.09
    Who
    -0.97
    Which
    -0.96
     What
    -0.94
     Where
    -0.79
    What
    -0.77
     Whose
    -0.75
    Where
    -0.74
     WHO
    -0.66
    POSITIVE LOGITS
     By
    1.52
    By
    1.49
     BY
    0.94
    BY
    0.77
     From
    0.77
    ByKey
    0.76
    ById
    0.72
    From
    0.71
    ByEmail
    0.69
    ByID
    0.68
    Activations Density 6.319%

    No Known Activations