© 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. 34616
    Prev
    Next
    INDEX
    Explanations

    increase

    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
    ads
    -0.09
     Rash
    -0.09
    ä¸į失
    -0.08
     Dud
    -0.08
    upo
    -0.08
     ras
    -0.08
    èģĶ
    -0.08
     Ñģли
    -0.08
    коп
    -0.08
    itness
    -0.07
    POSITIVE LOGITS
    人们çļĦ
    0.11
    åħ¬ç¤¾
    0.10
    ç¨ĭ度
    0.10
    åī§ç»Ħ
    0.10
    ä¸ļä½Ļ
    0.09
    æķ£çĥŃ
    0.09
    æĪijåİ»
    0.09
    ç½Ħ
    0.09
    æĥħä¾£
    0.09
    èĻĶ
    0.09
    Activations Density 0.100%

    No Known Activations