© 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. 38777
    Prev
    Next
    INDEX
    Explanations

    death, die, rest, cliche, tiring

    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
    åĨĻä¸ĭ
    -0.10
    TURE
    -0.09
    çĻ¾ç§ij
    -0.08
    /questions
    -0.08
    å¼ĢéĺĶ
    -0.08
    edian
    -0.08
    å¥ĩå¦Ļ
    -0.08
    çijĻ
    -0.08
    交æµģ
    -0.08
     pure
    -0.08
    POSITIVE LOGITS
    滥çĶ¨
    0.13
    åıįå¤į
    0.12
    é¢ijç¹ģ
    0.11
     clich
    0.11
    é¢ij
    0.10
    çĥĤ
    0.10
    家åĸ»æĪ·æĻĵ
    0.10
    次æķ°
    0.10
     repet
    0.10
    å·²ç»ı被
    0.10
    Activations Density 0.173%

    No Known Activations