© 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. 50880
    Prev
    Next
    INDEX
    Explanations

    first time

    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
    å¯ĴåĨ·
    -0.09
    ÐĦ
    -0.09
     ult
    -0.08
    relude
    -0.08
     sandwiches
    -0.08
    æĹłæīĢ
    -0.08
    å¨ĥå¨ĥ
    -0.08
    è§Ĵ度çľĭ
    -0.08
    èįī
    -0.08
    ily
    -0.08
    POSITIVE LOGITS
     оказÑĭва
    0.10
    大家éĥ½åľ¨
    0.10
    缸èģļ
    0.09
    cation
    0.09
    ìĶ©
    0.08
    大家åı¯ä»¥
    0.08
    ãģĹãģ¦ãģĬãģı
    0.08
    raj
    0.08
    鹦
    0.08
    çł´ç¢İ
    0.08
    Activations Density 0.020%

    No Known Activations