© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50138
    Prev
    Next
    INDEX
    Explanations

    departure after speculation

    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
    çķĮçļĦ
    -0.09
    åĩłåįģå¹´
    -0.09
    çŀij
    -0.09
    _operand
    -0.09
     redis
    -0.09
    çķĮ
    -0.09
    åIJijå¾Ģ
    -0.09
    åıijèĤ²
    -0.08
    enton
    -0.08
    redis
    -0.08
    POSITIVE LOGITS
    èģĮåĬ¡
    0.14
    è¾ŀèģĮ
    0.13
    æĽ´æį¢
    0.12
    ä»»èģĮ
    0.11
    çļĦä¿¡ä»»
    0.11
    å¤ĦåĪĨ
    0.11
    ä»»æľŁ
    0.11
    离èģĮ
    0.11
     resign
    0.10
    è£ģåijĺ
    0.10
    Activations Density 0.077%

    No Known Activations