© 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. 16566
    Prev
    Next
    INDEX
    Explanations

    code discussions

    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
    ç³»
    -0.06
    ç³»çļĦ
    -0.05
     bir
    -0.05
     vs
    -0.05
    2
    -0.05
      ↵
    -0.05
     nature
    -0.05
     vic
    -0.05
    åģĩ设
    -0.04
    ä¸ĭæŀ¶
    -0.04
    POSITIVE LOGITS
     Lég
    0.07
    åĩłå£°
    0.07
    awner
    0.07
    aled
    0.07
    elj
    0.07
    abaj
    0.07
    iszt
    0.07
    馬æĭī
    0.07
    clare
    0.07
    ovn
    0.07
    Activations Density 0.005%

    No Known Activations