© 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. 3353
    Prev
    Next
    INDEX
    Explanations

    natural language and related suffixes

    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
     bbc
    -0.09
    çīĩåŃIJ
    -0.09
    è¿ĩçļĦ
    -0.09
    تاÙĨ
    -0.09
    èĦı
    -0.09
    沿海
    -0.08
    ingly
    -0.08
    å¾Ĺèµ·
    -0.08
    ormal
    -0.08
    åĬ«
    -0.08
    POSITIVE LOGITS
    istic
    0.15
    istically
    0.13
    ists
    0.12
    ities
    0.11
    mente
    0.11
    ISTIC
    0.11
    isten
    0.10
    ist
    0.10
    ìłģìľ¼ë¡ľ
    0.10
    izing
    0.10
    Activations Density 0.032%

    No Known Activations