© 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. 64352
    Prev
    Next
    INDEX
    Explanations

    Capitalized words

    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
    çĽ¸å¯¹
    -0.31
    çĽ¸å¯¹è¾ĥ
    -0.29
    è·ĭ
    -0.26
    åıijå±ķçļĦ
    -0.25
    éĿ¢åIJij
    -0.25
    levant
    -0.25
     appet
    -0.24
     PEN
    -0.24
    ensitive
    -0.23
    ropolitan
    -0.23
    POSITIVE LOGITS
    tbody
    0.24
    Relations
    0.24
    ugging
    0.24
    ocks
    0.23
     Relation
    0.23
    åĪ©äºļ
    0.23
    Ruby
    0.23
    /manual
    0.23
    åĬĵ
    0.23
    кÑĥÑģ
    0.22
    Activations Density 1.503%

    No Known Activations