© 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. 14075
    Prev
    Next
    INDEX
    Explanations

    context clues and terminology in technical documentation, system descriptions, especially related to air conditioning and circuits

    oai_token-act-pair · gemini-2.0-flash

    say "wh" 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
     that
    -2.19
    that
    -1.83
     bahwa
    -1.48
     dass
    -1.31
     bahawa
    -1.30
     rằng
    -1.29
    THAT
    -1.20
     że
    -1.20
     että
    -1.18
    That
    -1.06
    POSITIVE LOGITS
    WH
    0.52
     Which
    0.49
    zky
    0.46
    iprot
    0.46
     who
    0.45
     wh
    0.45
    quels
    0.45
    Who
    0.44
     Who
    0.44
     which
    0.43
    Activations Density 1.584%

    No Known Activations