© 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. 6677
    Prev
    Next
    INDEX
    Explanations

    section navigations

    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
    è¾¾å°Ķ
    -0.09
    erate
    -0.08
    èĢ³
    -0.08
    è¡¿
    -0.08
    èĢ³æľµ
    -0.08
    ear
    -0.08
    èĴĮ
    -0.08
    æĶ¾è¿ĩ
    -0.08
    ugu
    -0.08
    è§Ħç«ł
    -0.08
    POSITIVE LOGITS
    æĭ¥
    0.10
     tes
    0.09
    inator
    0.09
    ç½²
    0.08
    .TR
    0.08
    ãģĿãģ®ä»ĸ
    0.08
    ä»İ容
    0.08
    è¿Ń
    0.08
    éĿ¢ä¸Ĭ
    0.08
    peaker
    0.08
    Activations Density 0.062%

    No Known Activations