© 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. 1846
    Prev
    Next
    INDEX
    Explanations

    version of

    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
    ingly
    -0.09
    orrect
    -0.08
    acha
    -0.08
    аниÑĨ
    -0.08
    çıī
    -0.08
    éķ°
    -0.08
    9
    -0.08
    velope
    -0.08
    iest
    -0.07
    _wr
    -0.07
    POSITIVE LOGITS
    FACT
    0.10
    æ½ĩæ´Ĵ
    0.10
    INDER
    0.10
    nel
    0.09
    èĤ¡æĿĥæĬķèµĦ
    0.09
     kê
    0.09
    çİĦ
    0.09
    imus
    0.09
    骥
    0.09
    istan
    0.09
    Activations Density 0.041%

    No Known Activations