© 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. 10776
    Prev
    Next
    INDEX
    Explanations

    parts, pieces, half, segments

    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
    è£Ĩ
    -0.10
     ultra
    -0.09
    Born
    -0.09
     Born
    -0.09
    æĪ´
    -0.09
    æ¡¢
    -0.09
    æĭ³
    -0.09
    æīĭç»Ń
    -0.09
    born
    -0.09
    æ¯Ĵ
    -0.09
    POSITIVE LOGITS
    ç¢İçīĩ
    0.16
     fragments
    0.15
     pieces
    0.15
    ç¢İ
    0.14
     parts
    0.13
    çł´ç¢İ
    0.13
    å®Įæķ´
    0.13
    parts
    0.13
    -parts
    0.12
    pieces
    0.12
    Activations Density 0.120%

    No Known Activations