© 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. 63092
    Prev
    Next
    INDEX
    Explanations

    studies and names

    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
    stances
    -0.28
    AccessorType
    -0.26
    å½ĵæĪij们
    -0.25
    åŁł
    -0.25
    åĽ¾çĶ»
    -0.25
    SEX
    -0.24
    è¾ķ
    -0.24
    alers
    -0.24
    ãģŀ
    -0.24
    inating
    -0.24
    POSITIVE LOGITS
    ä¹ħ
    0.28
    ç¿»
    0.27
     msm
    0.26
     игÑĢ
    0.26
    езÑĥлÑĮÑĤ
    0.26
     hạn
    0.26
     DVDs
    0.25
     improv
    0.24
    åIJĦ项
    0.24
     coz
    0.24
    Activations Density 8.400%

    No Known Activations