© 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. 11471
    Prev
    Next
    INDEX
    Explanations

    back to

    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
    移
    -0.09
    itions
    -0.09
    综åIJĪåĪ©çĶ¨
    -0.09
    uja
    -0.09
    å£ij
    -0.08
    oba
    -0.08
    éĵ¤
    -0.08
     sandwich
    -0.08
    åŃ½
    -0.08
    äºĶ大
    -0.08
    POSITIVE LOGITS
    imens
    0.09
    çķĮéĿ¢
    0.09
    éĦ±
    0.09
    第ä¸ĢæĿ¡
    0.08
    æĪĸå°Ĩ
    0.08
     Whale
    0.08
    enger
    0.08
     bullets
    0.08
    çĻ¾å®¶
    0.08
    ä¸įå±ij
    0.08
    Activations Density 0.054%

    No Known Activations