© 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. 9846
    Prev
    Next
    INDEX
    Explanations

    part of its

    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
    oes
    -0.10
    æĬ½åıĸ
    -0.10
    æĬ½
    -0.10
    rios
    -0.09
    æĦıè§ģ
    -0.09
    èķī
    -0.09
    缸åħ³åĨħ容
    -0.09
     zob
    -0.09
    exas
    -0.09
    èģļ
    -0.09
    POSITIVE LOGITS
    åĩĨå¤ĩå·¥ä½ľ
    0.08
    ç»ĦæĪIJéĥ¨åĪĨ
    0.08
    亲åĪĩ
    0.08
    å®£ä¼łæ´»åĬ¨
    0.08
    æĤ¬
    0.08
     overall
    0.07
    inan
    0.07
    æŃ£å¸¸ä½¿çĶ¨
    0.07
    adin
    0.07
    çļĦæķ´ä½ĵ
    0.07
    Activations Density 0.107%

    No Known Activations