© 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. 62934
    Prev
    Next
    INDEX
    Explanations

    Ren followed by 'egade', 'aming', 'ald', 'iform', 'owned', 'é'

    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
    urs
    -0.10
    ½æķ°
    -0.10
    ìļĶ
    -0.09
    buster
    -0.09
    ÑĩаÑĤ
    -0.09
    éĻ²
    -0.09
    åĥļ
    -0.09
    éĸĵéģķãģĦ
    -0.08
     Decay
    -0.08
    inary
    -0.08
    POSITIVE LOGITS
    ê¹IJ
    0.11
    ning
    0.10
    æĸĹå¿Ĺ
    0.09
    ninger
    0.09
    çĿ¬
    0.09
    egade
    0.09
    ovation
    0.09
    ìĿµ
    0.08
    äºĨä¸įå°ij
    0.08
    earch
    0.08
    Activations Density 0.023%

    No Known Activations