© 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. 51436
    Prev
    Next
    INDEX
    Explanations

    date or punctuation marking the end

    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
    Ñģад
    -0.11
    æĥħ趣
    -0.10
    çī¹æĸ¯
    -0.09
    ease
    -0.09
    éro
    -0.09
    licable
    -0.08
     lÆ°á»Ľi
    -0.08
    agini
    -0.08
     Bren
    -0.08
     perk
    -0.08
    POSITIVE LOGITS
    antly
    0.10
    éķ¶
    0.09
    ä¸įæĩĤ
    0.09
    æĢ»èĢĮè¨Ģä¹ĭ
    0.09
    ifier
    0.09
    å°ijæķ°
    0.09
    èĦ²
    0.09
    çľĭä¸įåĪ°
    0.08
    å¼Ĥè®®
    0.08
    ified
    0.08
    Activations Density 0.092%

    No Known Activations