© 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. 38200
    Prev
    Next
    INDEX
    Explanations

    plunge, plummet, dive

    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
    ullah
    -0.10
    ually
    -0.10
    bie
    -0.09
    -minded
    -0.09
     Bir
    -0.09
    ulary
    -0.09
    ulla
    -0.08
    ãģ¹
    -0.08
    upo
    -0.08
    istra
    -0.08
    POSITIVE LOGITS
    æ¾Ī
    0.11
     into
    0.10
     depths
    0.10
    ockey
    0.10
    å¾Īæ·±
    0.09
    æ½Ń
    0.09
    æĿĨèıĮ
    0.09
    æ¶ķ
    0.09
    steen
    0.09
    Into
    0.09
    Activations Density 0.010%

    No Known Activations