© 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. 17588
    Prev
    Next
    INDEX
    Explanations

    ert or et suffix followed by word

    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
    -0.15
    holder
    -0.15
    hit
    -0.14
    holders
    -0.13
    halt
    -0.13
    hand
    -0.13
    ãģĹãģ¾ãģĦ
    -0.12
     воÑĤ
    -0.12
    ร
    -0.12
    ettes
    -0.12
    POSITIVE LOGITS
    ting
    0.35
    ted
    0.24
    tings
    0.22
    TING
    0.22
    ters
    0.21
    ypes
    0.21
    ãĤ¥
    0.20
    ype
    0.19
    tingham
    0.18
    YPE
    0.18
    Activations Density 2.334%

    No Known Activations