© 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. 33333
    Prev
    Next
    INDEX
    Explanations

    contractions like 'll, 's, 't

    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
     ‘
    -0.16
     ’
    -0.13
     “
    -0.12
     ”
    -0.12
     “âĢ¦
    -0.11
     “[
    -0.11
    ’ll
    -0.10
    ’Brien
    -0.10
    ’ve
    -0.10
    ’B
    -0.10
    POSITIVE LOGITS
    ally
    0.09
    'y
    0.09
    ropri
    0.09
    éĶ¡
    0.09
    à¹Į
    0.09
    't
    0.08
    ----------------------------------------------------------------------------
    0.08
    'es
    0.08
    'il
    0.08
     "
    0.08
    Activations Density 0.470%

    No Known Activations