© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36118
    Prev
    Next
    INDEX
    Explanations

    i am, i have, i would

    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
    s
    -0.12
    ÑĤоÑĢ
    -0.10
    l
    -0.10
    æ®Ĩ
    -0.10
    mente
    -0.09
    nya
    -0.09
    lie
    -0.09
    ¿
    -0.09
    r
    -0.09
    nd
    -0.09
    POSITIVE LOGITS
    0.13
    å¿ĥè¡Ģ
    0.10
    irty
    0.10
    zelf
    0.10
    è¾Ī
    0.10
    .imgur
    0.10
     iVar
    0.09
    ostream
    0.09
    зв
    0.09
     /*<<<
    0.09
    Activations Density 0.094%

    No Known Activations