© 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. 27892
    Prev
    Next
    INDEX
    Explanations

    and then a noun

    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
    娩
    -0.09
    -esque
    -0.08
    èĵ¬åĭĥ
    -0.08
    croll
    -0.08
    çļĦåľ°åĽ¾
    -0.08
    社ä¼ļç§ijåѦ
    -0.08
    æģļ
    -0.08
    ç»ıæµİåŃ¦å®¶
    -0.08
    OW
    -0.07
    çļĦåľŁåľ°
    -0.07
    POSITIVE LOGITS
    оÑģоб
    0.10
    boats
    0.09
    åī©ä¸ĭ
    0.09
    大çĽĺ
    0.09
    DES
    0.09
    обÑĢаз
    0.08
    èµĦ
    0.08
    âĬĥ
    0.08
    woods
    0.08
    redicate
    0.08
    Activations Density 0.443%

    No Known Activations