© 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. 61083
    Prev
    Next
    INDEX
    Explanations

    full spectrum of or range

    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
    pec
    -0.09
    ndl
    -0.09
    ModelState
    -0.09
    4
    -0.08
    PEC
    -0.08
    æĢİä¹Īæł·
    -0.08
    ä¼łè¯´
    -0.08
    ophon
    -0.08
    æį¢
    -0.08
    åĪ»æĦı
    -0.08
    POSITIVE LOGITS
    èĮĥåĽ´
    0.21
     range
    0.20
    èĮĥåĽ´åĨħ
    0.19
    (range
    0.18
    range
    0.18
    .range
    0.16
    Range
    0.16
    _range
    0.16
    	range
    0.16
    -range
    0.16
    Activations Density 0.104%

    No Known Activations