© 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. 52909
    Prev
    Next
    INDEX
    Explanations

    a, a través, a nivel, a lo

    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
    pedia
    -0.09
    اÙĦ
    -0.09
    aub
    -0.09
    建设
    -0.09
     Geb
    -0.08
    ÑģоÑĤ
    -0.08
    宣è¨Ģ
    -0.08
    irie
    -0.08
    åħ¬æĬ¥
    -0.08
    åĥı个
    -0.08
    POSITIVE LOGITS
     través
    0.16
    èĩ³æĸ¼
    0.11
    缮åīį为æŃ¢
    0.11
     partir
    0.10
    imuth
    0.10
    sterdam
    0.10
    venture
    0.10
    terior
    0.10
    portun
    0.10
    ropriate
    0.10
    Activations Density 0.041%

    No Known Activations