© 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. 32227
    Prev
    Next
    INDEX
    Explanations

    Visual Studio project configuration

    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
     IDC
    -0.09
    æıĴ
    -0.09
     Wizard
    -0.08
    ä¸įæ¸ħ
    -0.08
    æłĵ
    -0.08
    åĨħ马
    -0.08
     Destiny
    -0.08
    æŃ¥
    -0.08
    è·ĥ
    -0.07
    ãģ»ãģĭ
    -0.07
    POSITIVE LOGITS
    çĭ¸
    0.10
    éĽĨåĽ¢èĤ¡ä»½
    0.10
    æĭĴä¸į
    0.09
    å¥ģ
    0.09
    åĪ¸
    0.09
    illas
    0.09
    代åĬŀ
    0.09
     Grip
    0.08
    æĦıè¯Ĩå½¢æĢģ
    0.08
    asters
    0.08
    Activations Density 0.575%

    No Known Activations