© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22302
    Prev
    Next
    INDEX
    Explanations

    White

    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
    mando
    -0.10
    uele
    -0.10
    ä¸ľè¥¿
    -0.10
    ewidth
    -0.10
    ingly
    -0.09
    thest
    -0.09
    çĥŃè¡Ģ
    -0.09
    ย
    -0.09
    arium
    -0.09
    imes
    -0.09
    POSITIVE LOGITS
    ened
    0.12
    fulness
    0.10
    åĩłåIJį
    0.10
     supremacist
    0.09
    ys
    0.09
    å²Ń
    0.09
    listed
    0.09
    åĩłå®¶
    0.09
    enance
    0.09
    urs
    0.09
    Activations Density 0.018%

    No Known Activations