© 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. 39453
    Prev
    Next
    INDEX
    Explanations

    archaeology and Indiana Jones

    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
    _hop
    -0.09
    .Dropout
    -0.08
     Hop
    -0.08
    hop
    -0.08
    usher
    -0.08
     поб
    -0.08
    趸
    -0.08
     зай
    -0.08
    å¥Ķ
    -0.08
    æĬķ
    -0.08
    POSITIVE LOGITS
    èĢĥåı¤
    0.21
    åĩºåľŁ
    0.18
    æĸĩçī©
    0.17
     archae
    0.15
    åıijæİĺ
    0.15
     Dirt
    0.14
    æĮĸæİĺ
    0.14
     Archae
    0.14
     archaeological
    0.13
    çĶ°éĩİ
    0.12
    Activations Density 0.075%

    No Known Activations