© 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. 5827
    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
    held
    -0.33
    isphere
    -0.30
    holding
    -0.28
     Äijo
    -0.28
    alled
    -0.27
    ometer
    -0.27
    iedo
    -0.26
    hold
    -0.25
    éļ¾
    -0.25
    .persistent
    -0.25
    POSITIVE LOGITS
    otch
    0.30
    ãģ¾ãģļãģ¯
    0.27
    é«ĺæķĪçļĦ
    0.26
    otec
    0.25
    otate
    0.25
    è¾¹çķĮ
    0.25
    tec
    0.24
     fast
    0.24
    lox
    0.24
    éĻ£
    0.24
    Activations Density 0.175%

    No Known Activations

    This feature has no known activations.