© 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. 62586
    Prev
    Next
    INDEX
    Explanations

    badly followed by adjective or verb

    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
    ed
    -0.09
    ackers
    -0.08
    çĻ¾åĪĨçĻ¾
    -0.08
    arie
    -0.08
    iom
    -0.08
    ï¸ı
    -0.08
    èĩªå¦Ĥ
    -0.08
    迳
    -0.07
    {"
    -0.07
    ly
    -0.07
    POSITIVE LOGITS
    inspace
    0.11
    è¿Ļæł·çļĦè¯Ŀ
    0.11
    entifier
    0.11
    ihu
    0.10
    éĩįå¿ĥ
    0.10
    styleType
    0.10
     seiz
    0.10
    schlie
    0.09
    åĵī
    0.09
    /workspace
    0.09
    Activations Density 0.007%

    No Known Activations