© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8576
    Prev
    Next
    INDEX
    Explanations

    demonstrated a prototype

    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
    åıijè´¢
    -0.11
    åıijå±ķ壮大
    -0.09
    æį¡
    -0.08
    avar
    -0.08
    çŁ¥éģĵèĩªå·±
    -0.08
    è´¢åĬ¡ç®¡çIJĨ
    -0.08
    aten
    -0.08
    ProcessEvent
    -0.08
    éĩįæķ´
    -0.08
    ocide
    -0.08
    POSITIVE LOGITS
    æ¼Ķ示
    0.17
    æµĭè¯ķ
    0.17
    ä½ĵéªĮ
    0.16
    示èĮĥ
    0.15
    è¯ķ
    0.15
     demos
    0.14
     ÑĤеÑģÑĤ
    0.14
    é«Ķé©Ĺ
    0.13
     test
    0.13
     testing
    0.13
    Activations Density 0.196%

    No Known Activations