© 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. 18536
    Prev
    Next
    INDEX
    Explanations

    technical and formal language

    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
     Stroke
    -0.31
    -strokes
    -0.30
     strokes
    -0.28
    Stroke
    -0.28
    rame
    -0.28
     stroke
    -0.26
    Administration
    -0.26
    æµĴ
    -0.26
     Nir
    -0.25
     Pulitzer
    -0.25
    POSITIVE LOGITS
    ighter
    0.29
    åľ°æĸ¹
    0.26
    udad
    0.25
    çŀ¬
    0.25
    wa
    0.25
    Wa
    0.25
     ngũ
    0.24
    LING
    0.24
    çĶ¸
    0.24
    大åĬĽ
    0.24
    Activations Density 0.831%

    No Known Activations