© 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. 35783
    Prev
    Next
    INDEX
    Explanations

    tuning or sounding

    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
    enen
    -0.10
    ion
    -0.10
    nt
    -0.09
    ðŁı»
    -0.09
    ives
    -0.09
     Jane
    -0.09
    ocht
    -0.08
    documentation
    -0.08
    ooter
    -0.08
     ÑĥÑĩаÑģÑĤ
    -0.08
    POSITIVE LOGITS
    STALL
    0.10
    æĿIJ
    0.10
    åŃIJéĩĮ
    0.09
    mith
    0.09
    NEL
    0.09
    æ°ijå¿ĥ
    0.09
    æĺī
    0.09
    ORITY
    0.09
    ipers
    0.09
    anca
    0.09
    Activations Density 0.011%

    No Known Activations