© 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. 2147
    Prev
    Next
    INDEX
    Explanations

    decimal numbers or words ending in d

    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
    æį»
    -0.09
    æįİ
    -0.09
    ney
    -0.08
    ðŁı»
    -0.08
    æĵĤ
    -0.08
    ies
    -0.08
    леп
    -0.08
    çĤ¹éĴŁ
    -0.08
    itus
    -0.08
    zer
    -0.08
    POSITIVE LOGITS
    eltas
    0.11
    ãĤ¥
    0.10
    entifier
    0.10
    伦
    0.10
    azed
    0.09
    greg
    0.09
    eref
    0.09
    alon
    0.09
    ensed
    0.09
    åĵī
    0.09
    Activations Density 0.114%

    No Known Activations