© 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. 11365
    Prev
    Next
    INDEX
    Explanations

    it followed by was, would, might

    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
    æĪijçļĦå¿ĥ
    -0.12
    æĪij认为
    -0.09
    顺
    -0.09
    æĪijå¿ĥéĩĮ
    -0.09
    åĪ°åºķ
    -0.09
    ç±³å°Ķ
    -0.09
    èĩªèº«çļĦ
    -0.08
    åĨ³å¿ĥ
    -0.08
    å¾Ģå¾Ģ
    -0.08
    ST
    -0.08
    POSITIVE LOGITS
    æľīå¿ħè¦ģ
    0.10
    æĹ©æĹ¥
    0.10
    æľºä¼ļ
    0.10
    ÑĢÑİ
    0.09
    elves
    0.09
    angen
    0.09
    osas
    0.09
    achu
    0.09
    _amp
    0.09
    нÑı
    0.09
    Activations Density 0.044%

    No Known Activations