© 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. 26801
    Prev
    Next
    INDEX
    Explanations

    Lanza, Weinstein, Moon, Gronkowski

    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
    лож
    -0.10
    è·ĥ
    -0.10
    æĻļ
    -0.09
    èĽ¸
    -0.09
    è¶ĭåĬ¿
    -0.09
    åģ¿
    -0.09
    å¾Ģ
    -0.09
    rying
    -0.09
    æĸ¯å¡Ķ
    -0.09
    èľĢ
    -0.08
    POSITIVE LOGITS
    ç»ĪçĶŁ
    0.10
    çĬ¯
    0.09
    argin
    0.09
    åĮºåĪ«
    0.09
    åıĮåıĮ
    0.09
    æĹģ
    0.08
    车éģĵ
    0.08
    InputDialog
    0.08
    ç¥Ń
    0.08
    GRAM
    0.08
    Activations Density 0.017%

    No Known Activations