© 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. 53019
    Prev
    Next
    INDEX
    Explanations

    View

    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
    æIJºå¸¦
    -0.29
    ä»ĸä¹Ł
    -0.26
    _Begin
    -0.25
    ISA
    -0.24
    åıijèµ·
    -0.24
    HAV
    -0.24
    -FIRST
    -0.24
    ä¿İ
    -0.24
    ener
    -0.24
     surviv
    -0.24
    POSITIVE LOGITS
    eron
    0.30
    er
    0.29
    eren
    0.27
     ved
    0.27
    ers
    0.27
    ãĥ³ãĤ°
    0.26
     Green
    0.26
     up
    0.25
    altern
    0.25
    olding
    0.25
    Activations Density 0.072%

    No Known Activations