© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9560
    Prev
    Next
    INDEX
    Explanations

    anger and related emotions

    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
     comprom
    -0.09
    èłķ
    -0.09
    _dl
    -0.09
    æģĭ
    -0.09
    çļĦ人çĶŁ
    -0.08
    vt
    -0.08
    èĵĸ
    -0.08
    VT
    -0.08
    dl
    -0.08
     dl
    -0.08
    POSITIVE LOGITS
    æĢĴ
    0.24
    æĦ¤æĢĴ
    0.23
    çĶŁæ°Ķ
    0.20
    æĦ¤
    0.17
    å¿¿
    0.17
    æĥħ绪
    0.16
     anger
    0.16
    anger
    0.15
    èĦ¾æ°Ķ
    0.15
     giáºŃn
    0.14
    Activations Density 0.070%

    No Known Activations