© 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. 4895
    Prev
    Next
    INDEX
    Explanations

    someone will help

    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
    usp
    -0.09
    iras
    -0.09
    obuf
    -0.09
    çļĦ妻åŃIJ
    -0.08
    妻
    -0.08
    é¬ĥ
    -0.08
    èĤīä½ĵ
    -0.08
    eton
    -0.08
    èĺij
    -0.08
    feof
    -0.08
    POSITIVE LOGITS
    æľī人
    0.16
     someone
    0.16
     somebody
    0.14
    Someone
    0.14
    someone
    0.13
     Someone
    0.13
    æľīä¸Ģä½į
    0.13
     Somebody
    0.12
     alguien
    0.12
     alguém
    0.11
    Activations Density 0.066%

    No Known Activations