© 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. 19249
    Prev
    Next
    INDEX
    Explanations

    Code and data

    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
    angu
    -0.29
     hur
    -0.27
    åħ¸
    -0.25
    corev
    -0.25
     gul
    -0.25
    òng
    -0.25
    AMI
    -0.25
    å®ŀ践活åĬ¨
    -0.25
    è¯ī
    -0.25
     Carm
    -0.25
    POSITIVE LOGITS
    stein
    0.28
    fred
    0.26
    filer
    0.26
    è¿Ļæĸ¹éĿ¢
    0.26
     Bird
    0.26
    lation
    0.26
    stration
    0.25
    å¼Ģä¸ļ
    0.25
    led
    0.25
    igen
    0.25
    Activations Density 0.010%

    No Known Activations