© 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. 18987
    Prev
    Next
    INDEX
    Explanations

    MIT License (MIT) Copyright (c)

    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
    ((
    -0.12
    (!
    -0.11
    çŃīåĪ°
    -0.10
    theless
    -0.10
    éĤ£éº¼
    -0.10
    (_,
    -0.09
    ï¼Ī
    -0.09
    æİ¥èijĹ
    -0.09
    ï¼Īç¬ij
    -0.09
    äºĨä»ĸçļĦ
    -0.09
    POSITIVE LOGITS
    или
    0.13
    i
    0.13
    aka
    0.13
    a
    0.11
    ie
    0.11
    以ä¸ĭç®Ģ称
    0.11
     aka
    0.11
    ...)↵
    0.10
    主è¦ģæĺ¯
    0.10
    åıĪ称
    0.10
    Activations Density 0.876%

    No Known Activations