© 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. 19292
    Prev
    Next
    INDEX
    Explanations

    benefits

    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
    enna
    -0.26
     mel
    -0.26
    ston
    -0.24
    éĢł
    -0.24
    kit
    -0.24
     Stub
    -0.24
    åķ¡
    -0.24
    .proxy
    -0.24
    feit
    -0.23
    åĨħ容
    -0.23
    POSITIVE LOGITS
     benefits
    0.38
    benef
    0.34
    working
    0.31
     working
    0.31
     Benefits
    0.30
    Benefits
    0.29
    çĽĬ
    0.28
    Ä©nh
    0.28
    Working
    0.28
     job
    0.27
    Activations Density 1.439%

    No Known Activations