© 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. 14525
    Prev
    Next
    INDEX
    Explanations

    version number

    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
     technological
    -0.19
     barr
    -0.19
    ahir
    -0.19
    WB
    -0.18
     ju
    -0.18
     tore
    -0.18
    leaf
    -0.18
    EEEE
    -0.18
    brook
    -0.18
    æijĬ
    -0.18
    POSITIVE LOGITS
    (fin
    0.20
    修补
    0.20
    ä¹ĥèĩ³
    0.20
     *&
    0.19
     вой
    0.18
    bsub
    0.18
     subst
    0.18
    UserDefaults
    0.18
    çĶļèĩ³æĺ¯
    0.17
    çĶļèĩ³
    0.17
    Activations Density 0.067%

    No Known Activations