© 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. 23628
    Prev
    Next
    INDEX
    Explanations

    Barnes & Barnett

    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
    çĦ¶æĺ¯
    -0.12
    ally
    -0.10
    tern
    -0.09
    lys
    -0.09
    íĥĦ
    -0.09
    /from
    -0.09
     Airways
    -0.09
    ly
    -0.09
    otron
    -0.08
    åĻİ
    -0.08
    POSITIVE LOGITS
    ëį°
    0.12
    OrUpdate
    0.11
    esion
    0.11
    egment
    0.10
    erge
    0.10
    ÙĪØ®
    0.10
    yard
    0.09
    perature
    0.09
    able
    0.09
    åĤħ
    0.09
    Activations Density 0.009%

    No Known Activations