© 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. 61142
    Prev
    Next
    INDEX
    Explanations

    exactly what someone wants

    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
    æĸ°åĮº
    -0.09
    aware
    -0.08
    è¿Ļä¹Īä¹ħ
    -0.08
    ote
    -0.08
    çĽĪ
    -0.08
    æĸĵ
    -0.08
    详ç»ĨçļĦ
    -0.08
    æĬĬè¿ĻäºĽ
    -0.08
    çĤī
    -0.08
     Confidential
    -0.08
    POSITIVE LOGITS
    åħ¶æīĢ
    0.19
     desired
    0.18
    æīĢéľĢè¦ģ
    0.18
    desired
    0.17
    æīĢéľĢ
    0.16
    ä½łåĸľæ¬¢
    0.14
    æīĢéľĢè¦ģçļĦ
    0.14
    æĪij们éľĢè¦ģ
    0.13
    ä½łæĥ³
    0.13
    èĩªå·±åĸľæ¬¢
    0.13
    Activations Density 0.150%

    No Known Activations