© 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. 14487
    Prev
    Next
    INDEX
    Explanations

    it's hard to

    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
     allegedly
    -0.24
    ä¸įæĸŃåľ°
    -0.20
     supposedly
    -0.20
     reportedly
    -0.19
     erotisk
    -0.18
    åĬłç´§
    -0.18
    èĭ¥å¹²
    -0.17
    æį®è¯´
    -0.17
    æ¾Ī
    -0.17
    ä¸Ģå¾ĭ
    -0.17
    POSITIVE LOGITS
    çľĭåĩº
    0.24
     feel
    0.24
    æĦŁè§ī
    0.23
     feeling
    0.23
     feels
    0.22
    æĦŁè¦º
    0.21
    太å¤ļçļĦ
    0.20
     lắm
    0.20
     cảm
    0.20
    oyo
    0.20
    Activations Density 0.022%

    No Known Activations