© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36334
    Prev
    Next
    INDEX
    Explanations

    names starting with K, C, or L followed by certain sub-tokens

    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
     slight
    -0.11
    noon
    -0.11
    places
    -0.11
    lando
    -0.10
    abin
    -0.10
    ãĥ£
    -0.09
    eer
    -0.09
    İ
    -0.09
     glad
    -0.09
     зло
    -0.09
    POSITIVE LOGITS
    ively
    0.11
    веÑĢÑħ
    0.10
    æĻ°
    0.10
    ustering
    0.10
    еÑģ
    0.10
     ÃŃch
    0.10
    ìį¨
    0.09
    enburg
    0.09
    sic
    0.09
    ural
    0.09
    Activations Density 0.477%

    No Known Activations