© 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. 19356
    Prev
    Next
    INDEX
    Explanations

    minimum tick

    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
    æľĹ
    -0.09
    çļĦ家åºŃ
    -0.09
    èĩªåįij
    -0.09
    éŵ
    -0.09
    è±ļ
    -0.08
    å¾ĺå¾Ĭ
    -0.08
    åİĮ
    -0.08
    avian
    -0.08
    å±Ģ
    -0.08
    éĤij
    -0.08
    POSITIVE LOGITS
    éĢĢå¸Ĥ
    0.11
    交æĺĵ
    0.10
     DataView
    0.10
    ç¯ģ
    0.09
    交æĺĵæīĢ
    0.09
    æĹĦ
    0.09
     issu
    0.09
    éŁŃ
    0.09
    沪
    0.09
    è§ĦåĪĻ
    0.09
    Activations Density 0.071%

    No Known Activations