© 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. 50531
    Prev
    Next
    INDEX
    Explanations

    fun

    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
    æĺİæľĹ
    -0.08
    æĬĢæľ¯æ°´å¹³
    -0.08
    ÑĥÑī
    -0.08
    èĮ«èĮ«
    -0.08
    cott
    -0.08
    <br
    -0.08
     Lots
    -0.07
    ÙĦاÙĤ
    -0.07
    oran
    -0.07
    ains
    -0.07
    POSITIVE LOGITS
    niest
    0.14
    ctors
    0.12
    erals
    0.12
    nier
    0.12
    nels
    0.11
    filled
    0.11
    amentals
    0.11
    ereal
    0.11
    -loving
    0.10
    åįģ足
    0.10
    Activations Density 0.017%

    No Known Activations