© 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. 48998
    Prev
    Next
    INDEX
    Explanations

    film

    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
     каÑĩ
    -0.10
    åĭĥ
    -0.10
    hta
    -0.09
    tsy
    -0.09
    éľĩ
    -0.09
    æĥ®
    -0.09
    -describedby
    -0.09
     плоÑī
    -0.08
    èģĶæĥ³
    -0.08
    Comparable
    -0.08
    POSITIVE LOGITS
    è¿Ļéĥ¨çĶµå½±
    0.14
    çĶµå½±
    0.13
     movies
    0.13
     movie
    0.12
     ìĺģíĻĶ
    0.10
    movies
    0.10
    movie
    0.10
     Filme
    0.09
    ä¸Ģéĥ¨
    0.09
     films
    0.09
    Activations Density 0.298%

    No Known Activations