© 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. 36366
    Prev
    Next
    INDEX
    Explanations

    skill and learning curve

    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
    identity
    -0.08
     Doub
    -0.08
     scanned
    -0.08
    å¾Īéĩįè¦ģ
    -0.08
    obot
    -0.08
    owl
    -0.08
    .quality
    -0.08
    ä¸Ģ线
    -0.08
    jav
    -0.08
    麽
    -0.08
    POSITIVE LOGITS
    éĹ¨æ§Ľ
    0.18
    éļ¾åº¦
    0.13
    æĻ¦
    0.13
    éĻĮçĶŁ
    0.12
    èī°
    0.12
    éĻ¡
    0.11
    驾é©Ń
    0.11
    .Requires
    0.11
     requiring
    0.11
    èī°èĭ¦
    0.11
    Activations Density 0.116%

    No Known Activations