© 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. 10372
    Prev
    Next
    INDEX
    Explanations

    movement towards or entering

    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
    è·ĥ
    -0.10
    åĢ¾
    -0.09
    幸
    -0.09
     ê´Ģìĭ¬
    -0.08
    æŀ¶åŃIJ
    -0.08
    yu
    -0.08
    èIJ½
    -0.08
     fate
    -0.08
    çĽijçĿ£
    -0.08
    ilt
    -0.08
    POSITIVE LOGITS
    éĢ²åħ¥
    0.12
    è¿Ľåħ¥
    0.12
    带åĪ°
    0.11
    apolis
    0.10
    è¿Ľåħ¥åĪ°
    0.10
    æĨ¨
    0.09
    ترÙĪ
    0.09
    Entering
    0.09
    _into
    0.09
    èī¯æĢ§
    0.09
    Activations Density 0.065%

    No Known Activations