© 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. 11559
    Prev
    Next
    INDEX
    Explanations

    Bridge, Alternative choice

    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
     professionally
    -0.25
     “
    -0.24
     reportedly
    -0.24
    åIJij社ä¼ļ
    -0.24
     professional
    -0.23
    engo
    -0.22
    professional
    -0.21
    orny
    -0.21
    éŃij
    -0.20
    ä¸ĵä¸ļçŁ¥è¯Ĩ
    -0.20
    POSITIVE LOGITS
    å©ł
    0.35
    çIJ¡
    0.33
    çijĶ
    0.31
    è¡ł
    0.31
    èĬł
    0.30
    çĪĶ
    0.30
     salopes
    0.30
    å©Į
    0.30
    æŵ
    0.30
    沺
    0.30
    Activations Density 0.042%

    No Known Activations