© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 4291
    Prev
    Next
    INDEX
    Explanations

    sky and above

    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
    cms
    -0.09
    è¸Ķ
    -0.08
    è¶ħ
    -0.08
    æĹłç¼ĺ
    -0.08
    饮
    -0.08
    ä½łä¸į
    -0.08
    itions
    -0.08
    æĹłå£°
    -0.08
    åĮIJ
    -0.08
    èĥ«
    -0.08
    POSITIVE LOGITS
    å¤©åľ°
    0.13
    åħīæĺİ
    0.13
    éĺ³åħī
    0.11
    æĺİåªļ
    0.11
     sunlight
    0.11
    åħīéĺ´
    0.11
     trá»Ŀi
    0.10
    衬
    0.10
    Anywhere
    0.10
    iali
    0.10
    Activations Density 0.027%

    No Known Activations