© 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. 24-TRANSCODER-HP
    7. 150123
    Prev
    Next
    INDEX
    Explanations

    blog snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Southampton
    -0.25
    æĢİä¹Īçľĭ
    -0.25
    .rad
    -0.25
     America
    -0.25
    brick
    -0.25
    ITT
    -0.25
    blr
    -0.24
    hydrate
    -0.24
    @update
    -0.24
    çĻĸ
    -0.24
    POSITIVE LOGITS
    æĬĢ
    0.26
    æĻ®æĥł
    0.26
     jeep
    0.25
    å¼Ģå±ķ
    0.25
    å®Ĺ
    0.25
     advoc
    0.25
    ongs
    0.25
    erce
    0.25
    uncated
    0.25
    hu
    0.25
    Activations Density 0.060%

    No Known Activations