© 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. 19391
    Prev
    Next
    INDEX
    Explanations

    may be, can be, could be

    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
    HING
    -0.07
    akest
    -0.06
     entweder
    -0.06
    zez
    -0.06
     Either
    -0.06
    adir
    -0.06
    Either
    -0.06
    ause
    -0.06
    好å¥ĩ
    -0.05
     either
    -0.05
    POSITIVE LOGITS
     ì¤Ħ
    0.07
    è¿ĩçļĦ
    0.06
    CLUDING
    0.06
    转è¿ĩ
    0.06
    åĨ³èµĽä¸Ń
    0.05
    anc
    0.05
    ijk
    0.05
     rou
    0.05
    combat
    0.05
    ظ
    0.05
    Activations Density 0.005%

    No Known Activations