© 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. 30-TRANSCODER-HP
    7. 95847
    Prev
    Next
    INDEX
    Explanations

    EPA and reporting

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     flow
    -0.37
     flows
    -0.35
    _flow
    -0.32
     Flow
    -0.30
    flow
    -0.30
    æµģ
    -0.29
     flowed
    -0.29
    .flow
    -0.28
     span
    -0.28
    çªľ
    -0.27
    POSITIVE LOGITS
    oho
    0.29
    ог
    0.29
    ogn
    0.26
     Observer
    0.26
    ograd
    0.26
    mul
    0.24
    illas
    0.24
    Obs
    0.24
    iba
    0.24
    åł±
    0.24
    Activations Density 0.004%

    No Known Activations