© 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. 63100
    Prev
    Next
    INDEX
    Explanations

    replication, validation, consistency

    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
     hữu
    -0.11
    _GAP
    -0.11
    RID
    -0.10
     СмоÑĤÑĢ
    -0.10
    å¾Īæĸ¹ä¾¿
    -0.09
     informed
    -0.09
    wright
    -0.09
    ÑĤал
    -0.09
    vae
    -0.08
    oll
    -0.08
    POSITIVE LOGITS
    æ£Ģ
    0.11
    红线
    0.10
    çīµ
    0.09
    好ä¸į容æĺĵ
    0.09
    红æĹĹ
    0.09
    éļ¾
    0.08
    åı¯ä¿¡
    0.08
    è¹ī
    0.08
    红èī²
    0.08
    éĽĴ
    0.08
    Activations Density 0.279%

    No Known Activations