© 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. 38371
    Prev
    Next
    INDEX
    Explanations

    derivative of

    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
    èĤĩ
    -0.08
    æļ´
    -0.08
     constit
    -0.08
    isen
    -0.08
    æĺ§
    -0.08
    èĥĦ
    -0.08
    为æĤ¨
    -0.08
     Bul
    -0.07
    iform
    -0.07
    è¿Ī
    -0.07
    POSITIVE LOGITS
    åħ¶å®ĥéĹ®é¢ĺ
    0.11
    è¿ŀç»Ń
    0.10
    åľ¨æĪijçļĦ
    0.10
    çĥŃæ°´åĻ¨
    0.10
    两年åīį
    0.09
    è¢į
    0.09
    éĢ£çºĮ
    0.09
    æ¡ĥ
    0.09
    undle
    0.09
    adol
    0.09
    Activations Density 0.288%

    No Known Activations