© 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. 15960
    Prev
    Next
    INDEX
    Explanations

    ending with ation or able

    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
    ically
    -0.29
    åŃIJéĩĮ
    -0.23
    isch
    -0.21
    Ùij
    -0.21
    иÑĩно
    -0.21
    çĦ¶æĺ¯
    -0.20
    åĦ¿
    -0.20
    ä¸įä¿¡
    -0.19
    icode
    -0.19
    iber
    -0.19
    POSITIVE LOGITS
    ation
    0.37
    ations
    0.36
    ment
    0.30
    ATION
    0.26
    è¿ĩçļĦ
    0.24
    ated
    0.24
    ator
    0.24
    acion
    0.24
    ance
    0.22
    able
    0.21
    Activations Density 0.424%

    No Known Activations