© 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. 3376
    Prev
    Next
    INDEX
    Explanations

    sense or feeling states

    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
     Devil
    -0.10
     Harden
    -0.09
    æıŃæĻĵ
    -0.08
    çĪ·
    -0.08
    åĵģåij³
    -0.08
    flt
    -0.08
    ç¬ı
    -0.08
    糣
    -0.08
     devil
    -0.08
    pei
    -0.08
    POSITIVE LOGITS
    æĦŁè§īèĩªå·±
    0.24
    è§īå¾Ĺèĩªå·±
    0.23
    æĦŁè§ī
    0.22
    æĦŁè§īåĪ°
    0.22
    ä¼ļè§īå¾Ĺ
    0.21
    è§īå¾Ĺ
    0.21
     feeling
    0.20
    æĦŁåĪ°
    0.20
     feel
    0.20
    覺å¾Ĺ
    0.19
    Activations Density 0.062%

    No Known Activations