© 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. 19-TRANSCODER-HP
    7. 8624
    Prev
    Next
    INDEX
    Explanations

    reward

    np_max-act · gemini-2.0-flash
    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
     stoi
    -0.25
    enders
    -0.25
    asion
    -0.25
    EO
    -0.25
    èļĬ
    -0.24
     vite
    -0.24
    ено
    -0.23
    穹
    -0.22
    eo
    -0.21
    åĨ³
    -0.21
    POSITIVE LOGITS
    spender
    0.25
    fully
    0.25
    -hour
    0.24
    /use
    0.23
    shr
    0.23
    满满
    0.23
    æĤ¸
    0.23
    æĮ¹
    0.23
    /result
    0.22
    -added
    0.21
    Activations Density 2.743%

    No Known Activations