© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 13976
    Prev
    Next
    INDEX
    Explanations

    Az + suffix

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    ing
    -0.20
    edly
    -0.17
    iveness
    -0.17
    i
    -0.16
    од
    -0.16
    zilla
    -0.15
    媳
    -0.15
    sr
    -0.15
    ÑĢÑĥ
    -0.15
    erer
    -0.14
    POSITIVE LOGITS
    0.22
    stra
    0.19
    imuth
    0.18
    oppel
    0.17
    ç»
    0.16
    Ķ
    0.16
    /OR
    0.15
    wins
    0.15
    Mex
    0.15
    meric
    0.14
    Activations Density 0.028%

    No Known Activations