© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21129
    Prev
    Next
    INDEX
    Explanations

    judged by jury or committee based on criteria

    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
    夹
    -0.11
    åįı
    -0.10
    /conf
    -0.09
    èĩªä¿¡
    -0.09
    è¯ķ
    -0.09
     Conf
    -0.09
    uju
    -0.08
    _guess
    -0.08
    è·³åĩº
    -0.08
    Conf
    -0.08
    POSITIVE LOGITS
     criteria
    0.12
    criteria
    0.12
     Criteria
    0.11
    Criterion
    0.10
    æ¼ĶæĬĢ
    0.10
     criter
    0.10
    ç¢į
    0.10
    riteria
    0.10
    饥
    0.10
    åħ«åŃĹ
    0.09
    Activations Density 0.054%

    No Known Activations