© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12474
    Prev
    Next
    INDEX
    Explanations

    compressing or compression

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    uate
    -0.19
    ì§ľ
    -0.16
    hz
    -0.15
    à¸Ľà¸ı
    -0.15
    ernal
    -0.15
    /or
    -0.15
    alog
    -0.15
    hood
    -0.15
    ãģ¶
    -0.15
    rus
    -0.15
    POSITIVE LOGITS
    ingly
    0.25
    ions
    0.23
    ible
    0.21
    /stretch
    0.19
    ors
    0.19
    andra
    0.18
    achusetts
    0.18
    ensible
    0.17
    ional
    0.17
    alte
    0.17
    Activations Density 0.012%

    No Known Activations