© 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. 1126
    Prev
    Next
    INDEX
    Explanations

    trash and garbage

    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
    ãĢħ
    -0.10
    ings
    -0.09
    ually
    -0.09
     влаÑģÑĤи
    -0.08
    ä¸įéĶĻçļĦ
    -0.08
    å¹ħ度
    -0.08
    注æĦıåĬĽ
    -0.08
    ube
    -0.08
    ãģ¹
    -0.08
    rees
    -0.08
    POSITIVE LOGITS
    æİīäºĨ
    0.12
    çŃIJ
    0.10
    æ»ĵ
    0.10
    俵
    0.10
    pile
    0.10
    æİī
    0.10
    yards
    0.10
    bins
    0.10
    ç¯ĵ
    0.09
     disposal
    0.09
    Activations Density 0.027%

    No Known Activations