© 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. 5852
    Prev
    Next
    INDEX
    Explanations

    signal to noise ratio

    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
    æ¶Ĥ
    -0.10
    ight
    -0.09
    uning
    -0.08
    墨西åĵ¥
    -0.08
    éķ¶
    -0.08
    åįĹéĿŀ
    -0.08
    åıįå¼¹
    -0.08
    æĭĨ
    -0.08
     Trailer
    -0.08
    åĪ¸åķĨ
    -0.08
    POSITIVE LOGITS
    åĻª
    0.14
     noise
    0.14
    åĻªéŁ³
    0.13
    åĻªå£°
    0.13
    noise
    0.12
    èĥĮæĻ¯
    0.11
     background
    0.10
    Noise
    0.10
    ä¸įåľ¨
    0.10
    _noise
    0.10
    Activations Density 0.273%

    No Known Activations