© 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. 18113
    Prev
    Next
    INDEX
    Explanations

    seen them, give a, one of, has a, else I, never really, cycle-dependent, thermal grill

    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
    æľ¬èº«çļĦ
    -0.09
    ilos
    -0.08
    ä¸įåĨįæĺ¯
    -0.08
     Pil
    -0.08
    æľ¬èº«
    -0.08
    çĺ¦èº«
    -0.08
    /about
    -0.08
    çĶ¬
    -0.08
    æĻĭ
    -0.08
    apr
    -0.07
    POSITIVE LOGITS
    è¿Ļä¹Īå¤ļ
    0.15
    æľīè¿Ļæł·çļĦ
    0.13
    éĤ£ä¹Īå¤ļ
    0.13
    å¦ĤæѤ
    0.12
    ä¸ºä½ł
    0.12
    ãģĵãģĵãģ¾ãģ§
    0.10
    è¿Ļä¹Ī
    0.10
    è¿Ļä¹Īä¹ħ
    0.10
    缸åIJĮ
    0.10
    ä¸İæѤ
    0.10
    Activations Density 0.028%

    No Known Activations