© 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. 57967
    Prev
    Next
    INDEX
    Explanations

    tip and tips

    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
     venda
    -0.09
    éĻħ
    -0.09
    ä¸Ģè¶Ł
    -0.08
    妻
    -0.08
    hed
    -0.08
    lide
    -0.08
    ampton
    -0.08
    eres
    -0.08
    latable
    -0.08
    iske
    -0.08
    POSITIVE LOGITS
    lessly
    0.14
    ologia
    0.12
    TURE
    0.12
    ologies
    0.11
    orary
    0.11
    pler
    0.10
    ðŁı»
    0.10
    ple
    0.10
    èĽĶ
    0.10
    usc
    0.10
    Activations Density 0.014%

    No Known Activations