© 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. 96289
    Prev
    Next
    INDEX
    Explanations

    Code snippets

    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
     sprung
    -0.30
    nte
    -0.27
     bl
    -0.27
    èĬĴ
    -0.25
    Foot
    -0.25
     collateral
    -0.25
    çģ«æĺŁ
    -0.25
    èĴ»
    -0.24
    å®ļäºĨ
    -0.24
    Bl
    -0.24
    POSITIVE LOGITS
    çĽ¥
    0.28
     mic
    0.27
     ROOM
    0.25
     Angle
    0.25
     nau
    0.25
    主ä½ĵ责任
    0.25
    OfString
    0.24
    ichage
    0.24
     Bond
    0.24
    çݯ
    0.24
    Activations Density 0.908%

    No Known Activations