© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8306
    Prev
    Next
    INDEX
    Explanations

    any [negative word]

    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
    orth
    -0.09
    .Raise
    -0.08
     recruited
    -0.08
     informant
    -0.08
    ree
    -0.08
    usher
    -0.08
    èĢĮæĺ¯
    -0.08
    è¿ĻéĩĮ
    -0.08
    ãĥ¥ãĥ¼
    -0.08
    åĽŀå½Ĵ
    -0.07
    POSITIVE LOGITS
     возможноÑģÑĤÑĮ
    0.13
    å·²ä¹ħçļĦ
    0.12
     possibility
    0.11
    æľīä»»ä½ķ
    0.11
    ä»»ä½ķå½¢å¼ı
    0.11
    以人为
    0.10
    åı¯èĥ½æĢ§
    0.10
     possibilitÃł
    0.09
    æĸĩåĮĸèĬĤ
    0.09
    èĢģå¦Ī
    0.09
    Activations Density 0.028%

    No Known Activations