© 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. 61825
    Prev
    Next
    INDEX
    Explanations

    copyright notice

    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
    0
    -0.10
     [âĢ¦]↵↵
    -0.10
    æ§
    -0.09
    çľ¼çļ®
    -0.09
    éĩijä»·
    -0.09
    cip
    -0.09
    æħ®
    -0.08
    esus
    -0.08
    פ
    -0.08
    èĿĮ
    -0.08
    POSITIVE LOGITS
    odore
    0.19
    etheless
    0.17
    gether
    0.14
    ÑįÑĤомÑĥ
    0.14
    adays
    0.12
    phalt
    0.11
    jourd
    0.11
    tlement
    0.11
    деÑģÑĮ
    0.10
    tempts
    0.10
    Activations Density 0.605%

    No Known Activations