© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 19197
    Prev
    Next
    INDEX
    Explanations

    Code, files, and paths

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     simulated
    -0.29
     sims
    -0.29
     lay
    -0.28
    就没æľī
    -0.27
    anian
    -0.27
    ä½Ī
    -0.26
     Sim
    -0.26
    _lineno
    -0.26
     simul
    -0.26
     Fires
    -0.25
    POSITIVE LOGITS
    _thumbnail
    0.28
    моÑĤ
    0.28
    -plan
    0.27
    _plan
    0.27
    .cv
    0.27
    pt
    0.27
    andle
    0.27
    plan
    0.27
    PT
    0.26
     capit
    0.26
    Activations Density 3.586%

    No Known Activations