© 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. 36061
    Prev
    Next
    INDEX
    Explanations

    size comparison

    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
    éĢĭ
    -0.09
    éĿ¢è²Į
    -0.09
    åĩŃè¯ģ
    -0.09
    amen
    -0.09
    人å±ħçݯå¢ĥ
    -0.09
    大桥
    -0.09
    åĪĨæľŁ
    -0.08
    æİīäºĨ
    -0.08
    çĶ¨èĩªå·±çļĦ
    -0.08
    大äºİ
    -0.08
    POSITIVE LOGITS
    -scale
    0.15
     amounts
    0.12
    å¹´å¤ľ
    0.11
    itarian
    0.11
    è¦ı模
    0.10
    elter
    0.10
    _endian
    0.10
    scale
    0.09
    éĵ¿
    0.09
     amount
    0.09
    Activations Density 0.159%

    No Known Activations