© 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. 52566
    Prev
    Next
    INDEX
    Explanations

    must be or should be

    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
    OUCH
    -0.10
    æĪĹ
    -0.09
    åĬŁåĬ³
    -0.09
     Beyond
    -0.08
     á»ķ
    -0.08
    æĪIJåĬŁçİĩ
    -0.08
    俱ä¹IJéĥ¨
    -0.08
    Beyond
    -0.08
     beyond
    -0.08
    åijĬè¯īä½ł
    -0.08
    POSITIVE LOGITS
    ç²¾éĢļ
    0.12
    èĥ½
    0.12
    åħ·æľī
    0.10
    æľīèĥ½åĬĽ
    0.10
     ability
    0.10
     able
    0.10
    çĨŁç»ĥ
    0.10
    ãģªãģĦãģ¨
    0.10
    çĶ¨å¾®ä¿¡
    0.10
    èĥ½å¤Ł
    0.10
    Activations Density 0.022%

    No Known Activations