© 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. 21762
    Prev
    Next
    INDEX
    Explanations

    divergence or diverged

    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
    over
    -0.10
    sv
    -0.09
    ackage
    -0.09
    äºĭåĦ¿
    -0.09
    空æ°Ķä¸Ń
    -0.08
    arov
    -0.08
    srv
    -0.08
    راس
    -0.08
    atical
    -0.08
     Buccane
    -0.08
    POSITIVE LOGITS
    iations
    0.11
    iator
    0.10
    iable
    0.10
    iation
    0.10
    æѧ
    0.10
    ius
    0.10
    iates
    0.09
    ly
    0.09
    gent
    0.09
    ients
    0.09
    Activations Density 0.020%

    No Known Activations