© 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. 37627
    Prev
    Next
    INDEX
    Explanations

    engine, range, capacity, performance

    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
    á»ģm
    -0.10
    é²ľæ´»
    -0.09
    èĮ¶
    -0.09
    æIJĶ
    -0.09
    çĶŁäº§è®¾å¤ĩ
    -0.09
    éĢıæĺİ
    -0.09
    涨价
    -0.09
    çĨ
    -0.09
    çļĦåģļæ³ķ
    -0.08
    çļĦå½±åĵį
    -0.08
    POSITIVE LOGITS
    å®īåħ¨æĢ§
    0.11
    转弯
    0.10
    è¡Į驶
    0.10
    稳å®ļæĢ§
    0.09
    éĢŁåº¦
    0.09
    æĢ§èĥ½
    0.09
    çģµæ´»æĢ§
    0.09
    æIJºå¸¦
    0.09
    çī¹çĤ¹æĺ¯
    0.08
    (single
    0.08
    Activations Density 0.057%

    No Known Activations