© 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. 58370
    Prev
    Next
    INDEX
    Explanations

    'us' followed by 'improve'

    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
    饥
    -0.12
    ä¼łåĩº
    -0.11
    沧
    -0.10
    _subtype
    -0.10
    æ¨Ł
    -0.09
    äºĪ
    -0.09
    订éĺħ
    -0.08
    COND
    -0.08
    enuity
    -0.08
    驾
    -0.08
    POSITIVE LOGITS
    ä»ħä¾ĽåıĤèĢĥ
    0.09
    æĦ¦
    0.09
    _NB
    0.08
    oni
    0.08
    ç½²
    0.08
     Deck
    0.08
    åIJİæĤĶ
    0.08
    追踪
    0.08
    ÑĢÑĥ
    0.07
    éĻªä½ł
    0.07
    Activations Density 0.070%

    No Known Activations