© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1542
    Prev
    Next
    INDEX
    Explanations

    from

    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
     metic
    -0.31
     CDDL
    -0.23
    ä»İåݻ年
    -0.22
    ä»İ
    -0.22
     XCTAssertTrue
    -0.21
    使
    -0.21
     XCTAssertEqual
    -0.20
     télécharg
    -0.20
     bureaucr
    -0.20
     perpetr
    -0.20
    POSITIVE LOGITS
    /to
    0.49
    /of
    0.31
    /by
    0.30
     afar
    0.29
    ä¸Ģå¼Ģå§ĭå°±
    0.29
    CharCode
    0.28
    ÅĤuż
    0.28
     whence
    0.27
    BOVE
    0.26
    _HERE
    0.24
    Activations Density 0.872%

    No Known Activations