© 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. 56511
    Prev
    Next
    INDEX
    Explanations

    must / find / somewhere / somehow

    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
    åĭ¿
    -0.09
    myp
    -0.08
    adh
    -0.08
    çĤĻ
    -0.08
    æĬĬæĪij
    -0.08
    ukt
    -0.08
    å¤ļéĩį
    -0.08
    满
    -0.08
    庸
    -0.08
    åıĮéĩį
    -0.08
    POSITIVE LOGITS
    å°±éľĢè¦ģ
    0.14
    å°±è¦ģ
    0.12
    å°±å¾Ĺ
    0.12
    ç»Ī究
    0.11
    ãģ©ãģĨãģĹãģ¦ãĤĤ
    0.11
    æĢ»æľī
    0.11
    å°±å¿ħé¡»
    0.11
    å°ijä¸įäºĨ
    0.11
    éĥ½ä¼ļæľī
    0.10
    æīįèĥ½
    0.10
    Activations Density 0.079%

    No Known Activations