© 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. 23289
    Prev
    Next
    INDEX
    Explanations

    training events and improvement

    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
    adge
    -0.10
    顾客
    -0.10
    èĽ¤
    -0.10
    é«ĵ
    -0.10
    _BUSY
    -0.10
    affle
    -0.10
    èĬ±çĶŁ
    -0.09
    elor
    -0.09
     Shed
    -0.09
    æĬ±åĽ¢
    -0.09
    POSITIVE LOGITS
    .train
    0.12
    (training
    0.11
    ä¸Ĭå²Ĺ
    0.10
     train
    0.10
    åħ»æĪIJ
    0.10
    íĽĪ
    0.09
    =train
    0.09
    [train
    0.09
    .training
    0.09
    è¿«
    0.09
    Activations Density 0.090%

    No Known Activations