© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37384
    Prev
    Next
    INDEX
    Explanations

    It is / also

    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
    åħ¶å¯¦
    -0.09
    ÙĪاس
    -0.09
    åħ¶å®ŀ
    -0.09
    ses
    -0.08
    è¿Ń
    -0.08
    yen
    -0.08
    ilen
    -0.08
     zwyk
    -0.08
    åħ¶å®ŀæĺ¯
    -0.08
    åĭĭ
    -0.08
    POSITIVE LOGITS
    ä¹ŁåĽłæѤ
    0.12
     ÑĤакже
    0.11
    è¿Ļä¹Ł
    0.10
    åIJĮæĹ¶ä¹Ł
    0.10
    åIJĮæŶè¿ĺ
    0.10
    éħĮ
    0.10
    éĢļè¿ĩ对
    0.09
     also
    0.09
    å¾Īå¿«å°±
    0.09
    ä»İèĢĮ使
    0.09
    Activations Density 0.132%

    No Known Activations