© 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. 7641
    Prev
    Next
    INDEX
    Explanations

    risk of

    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
    çĶŁçIJĨ
    -0.10
     pil
    -0.09
    ione
    -0.09
    å¨Ī
    -0.08
    mmc
    -0.08
     Pil
    -0.08
    çĻ½é©¬
    -0.08
    åıĤè°ĭ
    -0.08
     âĸij
    -0.08
    åħ¸åŀĭçļĦ
    -0.08
    POSITIVE LOGITS
    å±
    0.10
    çļĦçģµéŃĤ
    0.09
    oze
    0.09
    æµĨ
    0.09
    çļĦçĪ¶äº²
    0.09
    çĢļ
    0.09
    éģ¿
    0.09
    immers
    0.09
    çļĦçĶ·äºº
    0.08
    éķ¿å¯¿
    0.08
    Activations Density 0.477%

    No Known Activations