© 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. 6983
    Prev
    Next
    INDEX
    Explanations

    writing about

    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
    éĻĪè¿°
    -0.10
    宣è¨Ģ
    -0.09
    åŃ°
    -0.09
    UCT
    -0.09
    .getBean
    -0.09
    åĴİ
    -0.09
    ä»»ä½ķä¸Ģ个
    -0.09
    pery
    -0.09
    .IsValid
    -0.08
    IsValid
    -0.08
    POSITIVE LOGITS
    å¦Ĥä½ķ
    0.13
     how
    0.13
    亢
    0.11
    è¿ĻçĤ¹
    0.10
    å¦Ĥä½ķåİ»
    0.10
    æĺ¯å¦Ĥä½ķ
    0.10
    how
    0.10
     cómo
    0.09
     possibility
    0.09
    çļĦéĤ£ç§į
    0.09
    Activations Density 0.041%

    No Known Activations