© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 963
    Prev
    Next
    INDEX
    Explanations

    download

    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
    udd
    -0.09
    åĪ°æľŁ
    -0.09
    å°Ħ
    -0.09
    çļĦçľ¼
    -0.09
    uko
    -0.08
    ning
    -0.08
    lessly
    -0.08
    ỡ
    -0.08
    زÙĦ
    -0.08
    åı¨
    -0.08
    POSITIVE LOGITS
    able
    0.16
    ABLE
    0.12
    ables
    0.11
    éĵĥ
    0.10
    ì²ĺ
    0.10
    æѤåĪ»
    0.09
    sWith
    0.09
    OrCreate
    0.09
    ieren
    0.09
    å¤Ħ
    0.09
    Activations Density 0.043%

    No Known Activations