© 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. 38512
    Prev
    Next
    INDEX
    Explanations

    technical details

    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
    æ½į
    -0.10
    æIJģ
    -0.10
    æĶ¹è¿Ľ
    -0.10
    ippo
    -0.09
    éķ¶
    -0.09
    è¾¹
    -0.09
    æĶ¹èī¯
    -0.09
    æĤ±
    -0.09
    æĬĬæīĭ
    -0.09
    åĪĹ为
    -0.09
    POSITIVE LOGITS
    温æļĸ
    0.09
    çĥĺ
    0.09
    ätt
    0.08
    çİĦ
    0.08
    éŨæĪ·
    0.08
     dro
    0.08
    BUS
    0.08
     Fut
    0.08
    ht
    0.07
     presets
    0.07
    Activations Density 0.087%

    No Known Activations