© 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. 31-TRANSCODER-HP
    7. 18384
    Prev
    Next
    INDEX
    Explanations

    Code and foreign languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fare
    -0.27
     cod
    -0.27
    ibr
    -0.25
    ivers
    -0.25
    èŀĪ
    -0.25
    yll
    -0.24
    apl
    -0.24
    èħ°éĥ¨
    -0.24
    è¾ĵéĢģ
    -0.24
    planes
    -0.23
    POSITIVE LOGITS
    åĽ¢ä½ĵ
    0.25
     Zus
    0.24
    åįıè°ĥåıijå±ķ
    0.24
     group
    0.23
    群
    0.23
     groupName
    0.23
    æijĨ
    0.23
    .Sockets
    0.23
    ichtig
    0.23
     Group
    0.23
    Activations Density 0.418%

    No Known Activations