© 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. 2389
    Prev
    Next
    INDEX
    Explanations

    phrases starting with "I would guess/suggest/say/recommend"

    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
    èĥļ
    -0.09
    /GPL
    -0.09
     íijľ
    -0.09
    èķĥ
    -0.09
    åĤ¢
    -0.08
    mapper
    -0.08
    -sn
    -0.08
    asaki
    -0.08
    æŀ¶åŃIJ
    -0.08
    might
    -0.08
    POSITIVE LOGITS
    çĶ¨æĪ·æıIJä¾Ľ
    0.10
    éķĽ
    0.09
    æľĽ
    0.09
    é³ı
    0.08
    çİĻ
    0.08
    ä¾Ľå¤§å®¶
    0.08
    olly
    0.08
    åĨĴéĻ©
    0.08
    å²ļ
    0.08
    çĮ®
    0.08
    Activations Density 0.027%

    No Known Activations