© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8112
    Prev
    Next
    INDEX
    Explanations

    take over of

    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
    建åĨĽ
    -0.10
    tré
    -0.10
    tern
    -0.10
    emas
    -0.09
    端æŃ£
    -0.09
    .twig
    -0.09
    iments
    -0.09
    ivot
    -0.09
    ¼
    -0.09
    trl
    -0.09
    POSITIVE LOGITS
    éĺµåľ°
    0.08
    KeyType
    0.08
     pyl
    0.08
    驾é©Ń
    0.08
    åĴĮåıijå±ķ
    0.08
    åij½è¿IJ
    0.08
    и
    0.08
    æĵĤ
    0.08
    çİ°æľīçļĦ
    0.08
    åħļçļĦåįģä¹Ŀ
    0.08
    Activations Density 0.011%

    No Known Activations