© 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. 54106
    Prev
    Next
    INDEX
    Explanations

    legal principle or requirement

    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
    è¿Ŀ
    -0.08
    çĽħ
    -0.08
    dit
    -0.08
     mem
    -0.08
    æµģ泪
    -0.08
    alar
    -0.07
    enta
    -0.07
    ัà¸ģ
    -0.07
    htag
    -0.07
    rab
    -0.07
    POSITIVE LOGITS
    åħ¬å®īæľºåħ³
    0.09
    è¡ĮæĶ¿æľºåħ³
    0.09
    æľºåħ³
    0.09
    æ³ķå®ļ代表
    0.09
    足以
    0.09
    å·²çĦ¶
    0.09
     زÙħÙĨ
    0.08
    å¸Ĥåľºä¸»ä½ĵ
    0.08
     principalTable
    0.08
    OOSE
    0.08
    Activations Density 0.179%

    No Known Activations