© 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. 48168
    Prev
    Next
    INDEX
    Explanations

    in history

    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
    éĢĢä¼ij
    -0.09
     اÙĦÙĪاØŃد
    -0.08
    èĩ¾
    -0.08
     born
    -0.08
    çijĦ
    -0.08
     cum
    -0.08
    涵
    -0.08
    å²ģ以ä¸ĭ
    -0.08
    èİĬ
    -0.08
     подаÑĢ
    -0.08
    POSITIVE LOGITS
    åİĨåı²ä¸Ĭ
    0.23
     history
    0.21
    åı²ä¸Ĭ
    0.20
    åİĨåı²
    0.19
    history
    0.17
    åı²
    0.17
     History
    0.17
    .history
    0.15
    æŃ·åı²
    0.15
    _history
    0.15
    Activations Density 0.201%

    No Known Activations