© 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. 48432
    Prev
    Next
    INDEX
    Explanations

    instead 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
    åıªæĢķ
    -0.09
    åħ¬æŃ£
    -0.09
    åħħå®ŀ
    -0.08
     Ged
    -0.08
    åĴİ
    -0.08
    illance
    -0.08
    ç¨įç¨į
    -0.08
    æĬĢå·§
    -0.08
    å¾Ĺèµ·
    -0.08
    å¹³åĿĩæ°´å¹³
    -0.08
    POSITIVE LOGITS
     вмеÑģÑĤ
    0.16
     instead
    0.14
    误
    0.14
    éĶĻ误
    0.14
    éĶĻ
    0.13
    instead
    0.13
    Instead
    0.12
    éĶĻäºĨ
    0.12
     Instead
    0.11
    ä¸į该
    0.11
    Activations Density 0.401%

    No Known Activations