© 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. 11838
    Prev
    Next
    INDEX
    Explanations

    preceded by punctuation, often starting phrases

    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
    çĶļèĩ³æĺ¯
    -0.10
    jes
    -0.09
    æķ£
    -0.08
    åĪ«äººçļĦ
    -0.08
    æħ¢æħ¢
    -0.08
    env
    -0.08
    æĶ¶æķĽ
    -0.08
    å¾Īéķ¿
    -0.08
    çĶļèĩ³
    -0.08
    åĽłä¸ºæĪij
    -0.08
    POSITIVE LOGITS
    æį®ä»ĭç»į
    0.12
    æį®æĤī
    0.11
    æį®äºĨ解
    0.11
    æŃ¤æ¬¡æ´»åĬ¨
    0.10
    ãģªãģĬ
    0.09
    该项
    0.09
    ãģ¡ãģªãģ¿ãģ«
    0.09
    é«ĺåħ´
    0.08
    åİĨæŶ
    0.08
    é¦ĸ家
    0.08
    Activations Density 0.034%

    No Known Activations