© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50716
    Prev
    Next
    INDEX
    Explanations

    spin-off

    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
    ection
    -0.09
    ds
    -0.09
    çīĩåŃIJ
    -0.09
    符åIJĪæĿ¡ä»¶
    -0.08
    going
    -0.08
    agement
    -0.08
    	TokenName
    -0.08
    Misc
    -0.08
     Silva
    -0.08
    åı£æ°Ķ
    -0.08
    POSITIVE LOGITS
    ners
    0.13
    ning
    0.13
    NING
    0.11
    åijIJ
    0.10
    åĦ¿
    0.10
    aries
    0.09
    avigator
    0.09
    mö
    0.09
    å±ij
    0.09
    ally
    0.09
    Activations Density 0.019%

    No Known Activations