© 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. 58060
    Prev
    Next
    INDEX
    Explanations

    question marks after words

    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
     ??
    -0.17
     (?)
    -0.15
     ???
    -0.15
     ?????
    -0.14
     (?
    -0.13
    ?s
    -0.13
    (?
    -0.12
    ??
    -0.12
     {?
    -0.11
     ?
    -0.10
    POSITIVE LOGITS
    åij¢
    0.19
    ï¼Ł↵
    0.18
    ØŁ
    0.17
    ï¼Ł↵↵
    0.17
    Õŀ
    0.16
    ãģ§ãģĻãģĭ
    0.15
    ï¼Ł
    0.14
    ï¼Ł”
    0.13
    ï¼ŁãĢį
    0.13
    âĿĵ
    0.12
    Activations Density 0.641%

    No Known Activations