© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52832
    Prev
    Next
    INDEX
    Explanations

    line

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    OperationException
    -0.10
    /fw
    -0.10
    槿
    -0.09
    _pas
    -0.09
    ramework
    -0.09
    _Framework
    -0.08
    strup
    -0.08
    รว
    -0.08
    /runtime
    -0.08
    ARA
    -0.08
    POSITIVE LOGITS
    (Line
    0.15
    lined
    0.14
    -line
    0.13
    .getLine
    0.13
     Lines
    0.13
     lines
    0.13
    -Line
    0.13
    #line
    0.12
    .Line
    0.12
     Line
    0.12
    Activations Density 0.135%

    No Known Activations