© 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. 19-TRANSCODER-HP
    7. 7680
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åħ¶å®ĥéĹ®é¢ĺ
    -0.28
    åħ¶å®ĥ
    -0.24
    etter
    -0.24
     °
    -0.23
    请éĢīæĭ©
    -0.23
    cho
    -0.23
     Airways
    -0.23
    decision
    -0.23
    群
    -0.23
    rical
    -0.22
    POSITIVE LOGITS
    hw
    0.27
    hard
    0.27
    reno
    0.25
    士
    0.25
     Straw
    0.23
    hurst
    0.23
    软
    0.23
    èĻı
    0.23
     Abr
    0.23
     Hard
    0.22
    Activations Density 0.310%

    No Known Activations