© 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. 21802
    Prev
    Next
    INDEX
    Explanations

    effective contributions precision decisions implement available

    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
    ãģıãĤĬ
    -0.10
     Perm
    -0.09
    ë¹Ļ
    -0.08
    cover
    -0.08
    帮åĬ©ä¼ģä¸ļ
    -0.08
    è£Ķ
    -0.08
    istrate
    -0.08
    åıĬæŶ
    -0.08
    niest
    -0.08
     christ
    -0.08
    POSITIVE LOGITS
    æķ¬
    0.09
     amen
    0.09
    æĦŁåıĹåĪ°
    0.09
    对èĩªå·±çļĦ
    0.08
     inj
    0.08
    éĺħåİĨ
    0.08
    apas
    0.08
    æĬĬèĩªå·±
    0.08
    æĺĨä»ij
    0.08
    ÑĪек
    0.08
    Activations Density 0.613%

    No Known Activations