© 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. 11737
    Prev
    Next
    INDEX
    Explanations

    you can

    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
    ä¹ĭæīĢ以
    -0.25
     certainly
    -0.24
     probably
    -0.21
     zwar
    -0.21
    å½ĵçĦ¶
    -0.20
     also
    -0.20
    ä¸ĢèĪ¬
    -0.20
     likely
    -0.20
     generally
    -0.19
    çĶļèĩ³
    -0.19
    POSITIVE LOGITS
    è¹Ī
    0.21
    ses
    0.21
    碰åĪ°
    0.20
    regs
    0.20
    åĨµ
    0.19
     midst
    0.19
     behalf
    0.19
    /if
    0.19
    éĢīæĭ
    0.19
    æŃ»åİ»
    0.18
    Activations Density 0.158%

    No Known Activations