© 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. 55607
    Prev
    Next
    INDEX
    Explanations

    heart

    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
    å´ĩ
    -0.10
    åħį责声æĺİ
    -0.09
     Blink
    -0.09
     sig
    -0.08
    èĤĩ
    -0.08
    æĭ¼
    -0.08
     Sig
    -0.08
     mono
    -0.08
    aram
    -0.08
    sse
    -0.08
    POSITIVE LOGITS
    人å¿ĥ
    0.20
    æĪijçļĦå¿ĥ
    0.20
    å¿ĥåºķ
    0.16
    æĪijå¿ĥéĩĮ
    0.16
    å¿ĥ头
    0.15
    çļĦå¿ĥ
    0.15
    å¿ĥçģµ
    0.15
    å¿ĥéĩĮ
    0.15
     heart
    0.13
     hearts
    0.13
    Activations Density 0.057%

    No Known Activations