© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48233
    Prev
    Next
    INDEX
    Explanations

    promises and commitments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    éĩĩçĶ¨äºĨ
    -0.10
    æıŃå¼Ģ
    -0.09
    èĩªèIJ¥
    -0.09
    ç»ıèIJ¥æ´»åĬ¨
    -0.09
    å°Ĩ以
    -0.09
    侦
    -0.09
    æĪijåıijçİ°
    -0.09
     Directorate
    -0.08
    ä½łä¼ļåıijçİ°
    -0.08
    è¿IJä½ľ
    -0.08
    POSITIVE LOGITS
    æī¿è¯º
    0.15
     promised
    0.15
     originally
    0.14
    å½ĵåĪĿ
    0.13
     promise
    0.13
    ãģ¯ãģļ
    0.13
     promises
    0.13
    åİŁæľ¬
    0.12
    çŃĶåºĶ
    0.12
    åĪĿè¡·
    0.12
    Activations Density 0.175%

    No Known Activations