© 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. 38862
    Prev
    Next
    INDEX
    Explanations

    snake venom or draft

    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
    nero
    -0.10
    ©
    -0.10
    ounded
    -0.09
    ugu
    -0.09
    æłī
    -0.08
    seek
    -0.08
    ner
    -0.08
    mdl
    -0.08
    indle
    -0.08
    fait
    -0.08
    POSITIVE LOGITS
    pit
    0.11
    iest
    0.10
    busters
    0.10
    yro
    0.09
    èĪħ
    0.09
    lashes
    0.09
    yaml
    0.09
    laces
    0.09
    ä½łæĢİä¹Ī
    0.09
    ÙĪاÙĦ
    0.08
    Activations Density 0.007%

    No Known Activations