© 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. 38731
    Prev
    Next
    INDEX
    Explanations

    scientific/formal language

    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
    r
    -0.39
    al
    -0.34
    rac
    -0.33
    ar
    -0.30
    rp
    -0.29
    an
    -0.29
    rub
    -0.29
     a
    -0.29
    am
    -0.28
     constituent
    -0.28
    POSITIVE LOGITS
    ory
    0.36
    oric
    0.34
    ories
    0.32
    è¯ĨåĪ«
    0.30
    esign
    0.29
    ery
    0.29
    oras
    0.29
    ritical
    0.29
    erview
    0.28
    orical
    0.28
    Activations Density 0.002%

    No Known Activations