© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7892
    Prev
    Next
    INDEX
    Explanations

    purpose or reason

    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
    å¼Ħ
    -0.09
    æīĢè°ĵçļĦ
    -0.08
    ä¸Ģèĩ´æĢ§
    -0.08
    rig
    -0.08
     Alam
    -0.08
    atically
    -0.08
    Pie
    -0.08
     Ùħست
    -0.08
    ç¿ķ
    -0.08
    udas
    -0.08
    POSITIVE LOGITS
    缮çļĦ
    0.16
    çIJĨçĶ±
    0.15
    çļĦ缮çļĦ
    0.14
     purpose
    0.14
    çļĦçIJĨçĶ±
    0.13
    åİŁåĽł
    0.13
     reasons
    0.13
     목ìłģ
    0.12
     reason
    0.12
     purposes
    0.12
    Activations Density 0.075%

    No Known Activations