© 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. 11899
    Prev
    Next
    INDEX
    Explanations

    64-bit

    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
    LING
    -0.11
    rub
    -0.10
    WER
    -0.09
    olley
    -0.09
    uate
    -0.08
    æIJĵ
    -0.08
    ube
    -0.08
    æļĿ
    -0.08
     expectancy
    -0.08
    dle
    -0.08
    POSITIVE LOGITS
    åĿį
    0.09
    å¡Į
    0.09
    çĪ·
    0.08
    大ä¼ļ
    0.08
    åİŁè°ħ
    0.08
    åıijå±ķ空éĹ´
    0.08
    åĽ·
    0.08
    OSP
    0.08
    é¦ĸåıij
    0.08
    ç²¾ç»Ĩ
    0.08
    Activations Density 0.070%

    No Known Activations