© 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. 33270
    Prev
    Next
    INDEX
    Explanations

    mistake or screwing up

    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
    çIJĨ
    -0.11
    çIJĨä¼ļ
    -0.10
    èī²æĥħ
    -0.09
    ware
    -0.08
    onen
    -0.08
    å¹Ĥ
    -0.08
    强çĥĪ
    -0.08
     Fi
    -0.08
    é«ĺæ°´å¹³
    -0.08
    ãģĿ
    -0.08
    POSITIVE LOGITS
    åģļå¾Ĺ
    0.17
    失误
    0.14
    åģļæ³ķ
    0.13
    è¿ĩéĶĻ
    0.12
     mistakes
    0.11
    çĬ¯
    0.11
     wrongdoing
    0.10
    åĴİ
    0.10
    è¹ij
    0.10
    çĬ¯è§Ħ
    0.10
    Activations Density 0.118%

    No Known Activations