© 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. 5432
    Prev
    Next
    INDEX
    Explanations

    all of a sudden

    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
    utes
    -0.10
    å½ĵå±Ģ
    -0.09
    æıĨ
    -0.09
    æ®ĭ
    -0.09
    isis
    -0.09
    宿
    -0.08
    èĢ½è¯¯
    -0.08
    eya
    -0.08
     Ev
    -0.08
     autof
    -0.08
    POSITIVE LOGITS
    çªģçĦ¶
    0.28
    忽çĦ¶
    0.21
     suddenly
    0.21
    忽
    0.19
     sudden
    0.19
    çĮĽçĦ¶
    0.15
    ä¸Ģå¤ľ
    0.14
     Suddenly
    0.14
    çªģ
    0.14
    çªģåıij
    0.14
    Activations Density 0.032%

    No Known Activations