© 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. 19641
    Prev
    Next
    INDEX
    Explanations

    closer or step closer

    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
    润
    -0.08
     Pru
    -0.08
    鸦
    -0.08
    ç°¡
    -0.08
    vir
    -0.08
    溢
    -0.08
    ç®Ģ
    -0.08
     Pron
    -0.08
     Bolt
    -0.08
    infos
    -0.08
    POSITIVE LOGITS
    ä¸ĢæŃ¥
    0.14
    è¿Īè¿Ľ
    0.13
    羣缸
    0.13
    åĩłæŃ¥
    0.12
    æľĿçĿĢ
    0.10
     clos
    0.10
    éĢ¼è¿ij
    0.10
    羣åĪĩ
    0.10
    è¿Īåĩº
    0.10
    æıŃå¼Ģ
    0.10
    Activations Density 0.030%

    No Known Activations