© 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. 6308
    Prev
    Next
    INDEX
    Explanations

    Harvey followed by names

    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
    Negative Logits
    zelf
    -0.10
    ities
    -0.10
    å½Ģ
    -0.09
    eras
    -0.09
    enheim
    -0.09
    mates
    -0.09
     Simone
    -0.09
    itor
    -0.09
    ially
    -0.09
    gie
    -0.08
    POSITIVE LOGITS
    ingly
    0.11
    äºĨä¸Ģä¸ĭ
    0.09
    onte
    0.09
    rame
    0.09
    ä½łå°±
    0.09
    æĭ®
    0.09
    Ñĩик
    0.08
    ously
    0.08
    kins
    0.08
    çĽį
    0.08
    Activations Density 0.009%

    No Known Activations