© 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. 15458
    Prev
    Next
    INDEX
    Explanations

    instead, rather

    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
     presumably
    -0.25
    åĩłä¹İ
    -0.20
    trs
    -0.20
     reportedly
    -0.20
    низ
    -0.20
    dra
    -0.19
    rema
    -0.19
    åĩłä¹İæīĢæľī
    -0.19
    é¢Ģ
    -0.19
    uje
    -0.18
    POSITIVE LOGITS
    èĢĮä¸įæĺ¯
    0.35
     instead
    0.32
     plutôt
    0.32
     rather
    0.30
    instead
    0.30
    rather
    0.29
    ëĿ¼ëıĦ
    0.28
    æŁIJç§į
    0.27
    äºĨ解ä¸Ģä¸ĭ
    0.27
    èĢĮéĿŀ
    0.27
    Activations Density 0.210%

    No Known Activations