© 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. 7508
    Prev
    Next
    INDEX
    Explanations

    all-caps headings

    oai_token-act-pair · gemini-2.0-flash

    Code or technical language

    np_max-act-logits · gemini-2.0-flash

    plural -s Method used: 3. Reason: top logits emphasize "s"/suffixed forms

    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_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
    S
    -1.70
    getMenuInflater
    -0.69
    T
    -0.57
    SIS
    -0.57
     S
    -0.54
    E
    -0.54
    SIE
    -0.53
    Ss
    -0.52
    N
    -0.51
    Geplaatst
    -0.51
    POSITIVE LOGITS
    s
    1.44
    ся
    0.93
    ی
    0.87
    stil
    0.76
    spa
    0.76
    saw
    0.73
    schen
    0.73
    spe
    0.71
    ship
    0.71
    sai
    0.71
    Activations Density 3.596%

    No Known Activations