© 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. 107899
    Prev
    Next
    INDEX
    Explanations

    family

    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
     plus
    -0.32
     Plus
    -0.31
     imper
    -0.27
     behind
    -0.27
    plus
    -0.26
    umont
    -0.26
     compartment
    -0.26
    çݯ
    -0.26
     Encyclopedia
    -0.26
     ”
    -0.25
    POSITIVE LOGITS
    -word
    0.28
    'n
    0.26
    .createTextNode
    0.26
    izi
    0.26
     rozp
    0.26
    synthesize
    0.25
    compose
    0.25
    rete
    0.25
    ä¸ĩ人
    0.24
    lasting
    0.24
    Activations Density 1.804%

    No Known Activations