© 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. 20-TRANSCODER-HP
    7. 142240
    Prev
    Next
    INDEX
    Explanations

    Causal language

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ORN
    -0.26
    ewhere
    -0.25
    pared
    -0.25
     anchors
    -0.24
     fool
    -0.24
    joint
    -0.24
    jiang
    -0.24
     addAction
    -0.24
     Scre
    -0.24
    atron
    -0.24
    POSITIVE LOGITS
    olest
    0.28
    éĻIJæŶ
    0.26
    eneral
    0.26
    èĦī
    0.24
    ä¹ħä¹ħ
    0.24
    Subsystem
    0.24
    æ¾Ħæ¸ħ
    0.24
    lod
    0.24
    bound
    0.24
    æĮģç»Ń
    0.24
    Activations Density 12.824%

    No Known Activations