© 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. 56252
    Prev
    Next
    INDEX
    Explanations

    BSD-style license

    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
    bert
    -0.12
     bé
    -0.11
    Ń
    -0.11
    books
    -0.10
    iban
    -0.10
     bell
    -0.10
    apy
    -0.10
    upe
    -0.10
    âĶģâĶģ
    -0.10
    âĶģ
    -0.10
    POSITIVE LOGITS
    jamin
    0.17
     nhiêu
    0.14
    ruptcy
    0.14
     Isles
    0.13
    camp
    0.12
    å¨ĺ
    0.12
    ะ
    0.12
    iful
    0.11
    celona
    0.11
    irectional
    0.11
    Activations Density 2.384%

    No Known Activations