© 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. 150889
    Prev
    Next
    INDEX
    Explanations

    Quality control and monitoring

    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
    æĮ½
    -0.28
    çıij
    -0.27
    çĤ¹å¤´
    -0.25
    _guess
    -0.25
    asses
    -0.24
    大人
    -0.23
    ä»Ģä¹Īéĥ½
    -0.23
    laÅŁ
    -0.23
     mắn
    -0.23
    atica
    -0.23
    POSITIVE LOGITS
    QUAL
    0.27
     Caf
    0.26
    ENGINE
    0.25
    abilia
    0.25
    fix
    0.25
    åĪ¶åº¦
    0.24
    èģĮæĿĥ
    0.24
    /legal
    0.24
     Toolkit
    0.24
    ongo
    0.24
    Activations Density 0.904%

    No Known Activations