© 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. 39434
    Prev
    Next
    INDEX
    Explanations

    So far have

    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
    ä¸Ģå¼Ģå§ĭ
    -0.12
    ä¸Ģå¼Ģå§ĭå°±
    -0.11
    ä¹ħäºĨ
    -0.10
    æľĢç»Ī
    -0.10
    äºĭåħĪ
    -0.10
    å½ĵæŶ
    -0.10
    é²Ľ
    -0.10
    å¤ļå¹´åīį
    -0.09
    å°Ĩè¿Ľä¸ĢæŃ¥
    -0.09
    åIJİåĨį
    -0.09
    POSITIVE LOGITS
    è¿ĺ没æľī
    0.12
    éĥ½ä¸įæĺ¯
    0.10
    è¿ĺæľª
    0.10
     haven
    0.09
    éĥ½æĺ¯
    0.09
    etus
    0.09
    å¹¶æľª
    0.08
    _PD
    0.08
    _batches
    0.08
    hin
    0.08
    Activations Density 0.059%

    No Known Activations