© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6868
    Prev
    Next
    INDEX
    Explanations

    identified as

    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
    ä¹ĭä½ľ
    -0.08
    award
    -0.08
    æ¥Ķ
    -0.08
    éĵij
    -0.08
     Architecture
    -0.08
    Award
    -0.08
    å¿ī
    -0.08
    _parms
    -0.08
    åģļ个
    -0.08
    åįĬ个
    -0.08
    POSITIVE LOGITS
    å§ĵåIJį
    0.15
    å§ĵ
    0.14
    åIJįåı«
    0.13
    æĺ¯ä¸ĢåIJį
    0.11
    ä¸ĢåIJį
    0.10
    æĺ¯ä¸Ģä½į
    0.10
    åĪĿæŃ¥
    0.09
     tên
    0.09
    çĻ»è®°
    0.09
    åij½åIJį为
    0.09
    Activations Density 0.060%

    No Known Activations