© 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. 58788
    Prev
    Next
    INDEX
    Explanations

    truly positive or superlative

    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
    åijĹ
    -0.11
    æĬµ
    -0.10
    欹
    -0.09
    irected
    -0.09
    sing
    -0.09
    澡
    -0.09
    ä¹Łæ¯Ķè¾ĥ
    -0.08
    åŁºæľ¬åİŁåĪĻ
    -0.08
    .tom
    -0.08
    å§Ķ
    -0.08
    POSITIVE LOGITS
     truly
    0.18
    羣æŃ£
    0.16
    羣æŃ£çļĦ
    0.14
     Truly
    0.13
    æīįæĺ¯çľŁæŃ£
    0.12
    æĦıä¹īä¸ĬçļĦ
    0.10
    好好
    0.10
     genuinely
    0.10
    å½»åºķ
    0.10
    åĬŁå¤«
    0.09
    Activations Density 0.043%

    No Known Activations