© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 18406
    Prev
    Next
    INDEX
    Explanations

    {% endfor %}

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    vor
    -0.05
    rait
    -0.05
    ÑĢо
    -0.05
    ilan
    -0.05
    人å¿ĥçļĦ
    -0.05
    以ä¸ĭæĿ¡ä»¶
    -0.05
    licit
    -0.05
    ÙĦات
    -0.05
    Ñĥж
    -0.05
    [:,
    -0.05
    POSITIVE LOGITS
     graved
    0.06
     possu
    0.06
     secretaria
    0.06
    dle
    0.06
     incentivo
    0.06
    føre
    0.06
    ом
    0.06
    presenta
    0.06
    upaya
    0.06
    èıĮ
    0.06
    Activations Density 0.010%

    No Known Activations