© 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. 10728
    Prev
    Next
    INDEX
    Explanations

    weak

    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
    wich
    -0.17
    añ
    -0.16
    vor
    -0.16
    uccess
    -0.16
    edo
    -0.15
    _PAIR
    -0.15
    åıijå±ķæľºéģĩ
    -0.15
    丰ç¡ķ
    -0.15
    ãģĨãģ¡ãģ«
    -0.15
    ready
    -0.15
    POSITIVE LOGITS
    ened
    0.25
     mẽ
    0.25
    ly
    0.24
    listed
    0.23
    à¹Ĩ
    0.23
    poke
    0.22
     à¹Ĩ
    0.22
    ingly
    0.21
    ens
    0.21
    lings
    0.20
    Activations Density 0.014%

    No Known Activations