© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 49059
    Prev
    Next
    INDEX
    Explanations

    rapid changes

    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
    olle
    -0.09
    ÑĪка
    -0.09
    liness
    -0.09
    pile
    -0.09
    paces
    -0.09
    ever
    -0.09
    éĤĪ
    -0.09
    ilton
    -0.08
    亲åıĭ
    -0.08
     Almighty
    -0.08
    POSITIVE LOGITS
    -fire
    0.13
    ity
    0.12
    ness
    0.11
     enough
    0.11
     جدا
    0.11
    çĽ¸ä¼ł
    0.10
    emente
    0.10
    iously
    0.10
    -response
    0.10
    edException
    0.10
    Activations Density 0.018%

    No Known Activations