© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 5398
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     suprem
    -0.30
    é¾Ħ
    -0.27
     lá»ĭch
    -0.26
     premise
    -0.26
     age
    -0.25
    achen
    -0.25
    heiten
    -0.25
     volume
    -0.25
     gain
    -0.24
     LOVE
    -0.24
    POSITIVE LOGITS
     Opport
    0.27
    æIJħ
    0.27
    寻常
    0.26
    utters
    0.26
    æ°
    0.25
     borderTop
    0.25
    åĪ»
    0.24
     afterward
    0.24
     hashed
    0.24
     wreck
    0.24
    Activations Density 0.021%

    No Known Activations

    This feature has no known activations.