© 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. 33432
    Prev
    Next
    INDEX
    Explanations

    good, bad, or positive evaluations

    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
    åħ¶å®ŀ
    -0.09
    è¿¢
    -0.09
     ìĦ¤
    -0.09
    ä¸Ģä¸ĭåŃIJ
    -0.09
    å¤ļå°ij
    -0.09
    éŀł
    -0.09
    æĹĹå¸ľ
    -0.08
    erval
    -0.08
    å¹ħ度
    -0.08
    еÑĩа
    -0.08
    POSITIVE LOGITS
    æŃ£éĿ¢
    0.26
     positive
    0.23
    positive
    0.20
    -positive
    0.19
    好çļĦ
    0.19
    Positive
    0.19
    好äºĭ
    0.18
     Positive
    0.18
    好è¯Ħ
    0.18
     positives
    0.17
    Activations Density 0.119%

    No Known Activations