© 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. 9767
    Prev
    Next
    INDEX
    Explanations

    time stamp or signature

    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
    åĪ°æŶåĢĻ
    -0.11
    ments
    -0.11
    ly
    -0.11
    ãģĦãģ¾ãģĻ
    -0.11
    ment
    -0.10
    ãģĦãģij
    -0.09
    nt
    -0.09
    ry
    -0.09
    ฯ
    -0.09
    æѤæŶ
    -0.09
    POSITIVE LOGITS
    ofday
    0.14
    fully
    0.14
    æĪ³
    0.13
    lessly
    0.13
    perature
    0.12
    -consuming
    0.12
    cales
    0.12
    othy
    0.12
    keeping
    0.12
    lessness
    0.11
    Activations Density 0.182%

    No Known Activations