© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9208
    Prev
    Next
    INDEX
    Explanations

    originated from, spread to, started from

    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
    èĢĹ
    -0.08
    -ons
    -0.08
    åĨī
    -0.08
    abbr
    -0.08
    itone
    -0.08
    itect
    -0.08
    ials
    -0.07
    çŃ¹å»º
    -0.07
    æŁ¬
    -0.07
    .viewport
    -0.07
    POSITIVE LOGITS
    åıijæĺİ
    0.16
    æľĢæĹ©çļĦ
    0.13
    æľĢæĹ©
    0.13
    èµ·æºIJ
    0.12
    欧洲
    0.12
    æºIJäºİ
    0.11
    æ¼Ķåıĺ
    0.11
     invention
    0.11
    æľĢåĪĿ
    0.11
    åĪĽæĸ°åıijå±ķ
    0.11
    Activations Density 0.157%

    No Known Activations