© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1304
    Prev
    Next
    INDEX
    Explanations

    factual reporting

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    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
     Gespr
    -0.39
     Geschä
    -0.38
    zcze
    -0.34
     Geschäfts
    -0.33
     Gründe
    -0.32
     persön
    -0.32
     tatsäch
    -0.31
     Führung
    -0.30
     Fälle
    -0.29
     völl
    -0.29
    POSITIVE LOGITS
    Nr
    0.39
     âĢŀ
    0.35
    _nr
    0.35
     hereby
    0.33
     splitted
    0.31
    âĢŀ
    0.30
     harbour
    0.30
     inflamm
    0.28
     nowadays
    0.28
     deberá
    0.28
    Activations Density 9.026%

    No Known Activations