© 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. 39865
    Prev
    Next
    INDEX
    Explanations

    naming and renaming specific things

    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
    éĽĨä¸Ńåľ¨
    -0.09
    APH
    -0.09
    çĪ±ä¸Ĭ
    -0.08
    æĢ¯
    -0.08
     destruct
    -0.08
    æīĭéĩĮ
    -0.08
     полÑĮзÑĥ
    -0.07
    èĩªèIJ¥
    -0.07
    带ç»Ļ
    -0.07
    ordes
    -0.07
    POSITIVE LOGITS
    åij½åIJį
    0.13
     naming
    0.13
    åij½åIJį为
    0.12
     named
    0.11
    æ°¸è¿ľ
    0.10
    éĻ¡
    0.10
    æĹłåĬ©
    0.09
    (named
    0.09
    esco
    0.09
     renaming
    0.09
    Activations Density 0.157%

    No Known Activations