© 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. Qwen3-4B
    5. 35-TRANSCODER-HP
    7. 40645
    Prev
    Next
    INDEX
    Explanations

    Rearranging sequences/equations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     ret
    -0.29
     ro
    -0.28
    ive
    -0.24
    Ro
    -0.23
    ROID
    -0.23
     RO
    -0.23
    .RegularExpressions
    -0.23
     anon
    -0.23
    imal
    -0.23
     row
    -0.23
    POSITIVE LOGITS
    r
    1.23
    rack
    0.59
    rz
    0.57
    rk
    0.56
    rms
    0.54
    rar
    0.53
    rng
    0.52
    rut
    0.51
    rug
    0.49
    rab
    0.48
    Activations Density 20.271%

    No Known Activations