© 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. 7646
    Prev
    Next
    INDEX
    Explanations

    out followed by very, score, team, or of

    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
    åºĭ
    -0.09
    æ³
    -0.09
     nid
    -0.09
    æĪ¿
    -0.08
    .populate
    -0.08
    éķĽ
    -0.08
    ÑĢем
    -0.08
    å¨ĥ
    -0.08
    èĭ¦
    -0.08
    mour
    -0.08
    POSITIVE LOGITS
     SON
    0.10
    -FIRST
    0.09
    æŃ£éĿ¢
    0.09
    çħ§
    0.09
    ampion
    0.08
    æĤ¬
    0.08
     Stall
    0.08
    _ghost
    0.08
    æĥ³åĪ°äºĨ
    0.08
    çģŀ
    0.08
    Activations Density 0.007%

    No Known Activations