© 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. 40771
    Prev
    Next
    INDEX
    Explanations

    u

    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
    top
    -0.44
    b
    -0.37
    t
    -0.35
    ris
    -0.31
    r
    -0.31
    rm
    -0.31
    tiles
    -0.31
    bic
    -0.31
    typ
    -0.29
    tl
    -0.29
    POSITIVE LOGITS
    xes
    0.31
    xor
    0.31
    gef
    0.28
    xde
    0.28
    xl
    0.27
    éĢļè¿ĩ
    0.27
    xon
    0.26
    gil
    0.26
    fan
    0.26
    xo
    0.25
    Activations Density 69.159%

    No Known Activations