© 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. 33-TRANSCODER-HP
    7. 53598
    Prev
    Next
    INDEX
    Explanations

    Color chemical analysis

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IJľ
    -0.29
    åİŁæłĩé¢ĺ
    -0.28
    ResourceId
    -0.26
     pt
    -0.26
    ç¯ĩ
    -0.25
    lide
    -0.24
     fringe
    -0.24
    sizeof
    -0.24
    yt
    -0.23
    chor
    -0.23
    POSITIVE LOGITS
     turno
    0.28
    çīĮ
    0.27
    obili
    0.25
     sadd
    0.25
    çľĭå¾Ĺ
    0.24
     cheer
    0.24
    rina
    0.24
    pdo
    0.23
    Pal
    0.23
    oso
    0.23
    Activations Density 0.026%

    No Known Activations