© 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. 38139
    Prev
    Next
    INDEX
    Explanations

    Star Trek, Star Wars, Starling, Starwood

    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
    ½æķ°
    -0.11
    绿水
    -0.09
    each
    -0.09
    ç»ĪçĶŁ
    -0.09
    æĭĮ
    -0.08
    abox
    -0.08
    _PULL
    -0.08
     jack
    -0.08
    代è¨Ģ
    -0.08
    ات
    -0.08
    POSITIVE LOGITS
    rett
    0.11
    ships
    0.10
    æĩ¿
    0.09
    backs
    0.09
    kest
    0.09
    light
    0.08
    renc
    0.08
    mites
    0.08
    listed
    0.08
    ship
    0.08
    Activations Density 0.011%

    No Known Activations