© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 22378
    Prev
    Next
    INDEX
    Explanations

    from ... to

    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
    Negative Logits
    å·´æĭī
    -0.07
     bras
    -0.06
    åĬ²
    -0.06
    è¸
    -0.05
    åĬ²åĦ¿
    -0.05
    abra
    -0.05
     spre
    -0.05
    emale
    -0.05
     cul
    -0.05
    IGO
    -0.05
    POSITIVE LOGITS
    åĮħ
    0.05
    æĻºèĥ½
    0.05
    ables
    0.05
     verkligen
    0.05
    è¿ĺ没æľī
    0.05
    çĦ¶å¤§
    0.05
    lop
    0.05
    åŃļ
    0.05
    ój
    0.05
    apta
    0.05
    Activations Density 0.119%

    No Known Activations