© 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. 39567
    Prev
    Next
    INDEX
    Explanations

    "is the" or "to be the"

    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
    spNet
    -0.10
    oldt
    -0.10
    ksz
    -0.09
    enny
    -0.09
    sPid
    -0.09
    åı«åģļ
    -0.09
    HeaderCode
    -0.09
    าว
    -0.09
    ONGL
    -0.09
    .Foundation
    -0.09
    POSITIVE LOGITS
    éĤ£æ®µ
    0.09
    åħ¶ä¸Ńä¸Ģ个
    0.09
    ä½Ī
    0.09
    èµĸ以
    0.08
    款
    0.08
    ä¸Ģ段
    0.08
     one
    0.08
    亹
    0.07
    å¸Ŀçİĭ
    0.07
    ube
    0.07
    Activations Density 0.193%

    No Known Activations