© 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. 32-TRANSCODER-HP
    7. 42261
    Prev
    Next
    INDEX
    Explanations

    Negation, conditionality, exception

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     either
    -0.36
    either
    -0.36
     Either
    -0.36
    Either
    -0.34
    yet
    -0.28
    ston
    -0.28
    bull
    -0.28
    åį´
    -0.26
     yet
    -0.26
    ä½Ĩåį´
    -0.26
    POSITIVE LOGITS
     ever
    0.39
     Ever
    0.33
     EVER
    0.32
    Ever
    0.28
     Orch
    0.27
     diffuse
    0.26
    å®ļåIJij
    0.26
    -ever
    0.25
    objc
    0.25
    èĩ´
    0.25
    Activations Density 0.005%

    No Known Activations