© 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. 62370
    Prev
    Next
    INDEX
    Explanations

    better world or place

    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
    è¿Ļæł·åģļ
    -0.09
    åħ¨æµģç¨ĭ
    -0.09
    å¾Ĺå¾Ī
    -0.08
    蹶
    -0.08
    å¸ĤåľºçļĦ
    -0.08
    å¾Īæ·±
    -0.08
    åħ¨ç¨ĭ
    -0.08
    è¯Ŀ说
    -0.08
    主å¸Ń
    -0.08
    æĥ³åĪ°
    -0.08
    POSITIVE LOGITS
    ç¾İ好çĶŁæ´»
    0.12
    æĶ¹åĸĦ
    0.09
     Better
    0.09
    ÑĤÑıж
    0.09
    å®ľ
    0.09
     better
    0.09
    æĬµè¾¾
    0.09
    intr
    0.09
    ä¹ĥ
    0.09
    沤
    0.09
    Activations Density 0.042%

    No Known Activations