© 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. 54892
    Prev
    Next
    INDEX
    Explanations

    after 'then' or 'when'

    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
    åIJ¸
    -0.10
    etti
    -0.09
    ä¸į失
    -0.09
    ç»ĻäºĨæĪij
    -0.09
    çľģæĶ¿åºľ
    -0.09
    èŀį为ä¸Ģä½ĵ
    -0.09
    å®īæħ°
    -0.09
    ãģ¡ãĤĥãĤĵãģ¨
    -0.09
    _STARTED
    -0.08
    åIJģ
    -0.08
    POSITIVE LOGITS
    飫
    0.10
    å°±æľīäºĨ
    0.09
     reportedly
    0.09
    TypeEnum
    0.09
    éħ¿
    0.08
    æĮģæľī
    0.08
    èĩªå¸¦
    0.08
    å½ķéŁ³
    0.08
    æį®
    0.08
    æį®äºĨ解
    0.08
    Activations Density 0.080%

    No Known Activations