© 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. 26945
    Prev
    Next
    INDEX
    Explanations

    starting point

    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
    .throw
    -0.10
    éļĶ
    -0.09
    æĽ´å¤ļ
    -0.09
    åµĺ
    -0.09
    彬
    -0.09
    æīĺ
    -0.09
    rlen
    -0.09
    isel
    -0.09
    骨
    -0.09
    _Desc
    -0.08
    POSITIVE LOGITS
    ä¸ĭæīĭ
    0.15
    starting
    0.14
    èµ·æŃ¥
    0.13
    èµ·çĤ¹
    0.13
     starting
    0.13
    åħ¥æīĭ
    0.12
    å§ĭ
    0.12
     Starting
    0.12
    åĪĩåħ¥çĤ¹
    0.12
    Starting
    0.12
    Activations Density 0.106%

    No Known Activations