© 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. 42452
    Prev
    Next
    INDEX
    Explanations

    right

    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
     required
    -0.31
    å¿ħ
    -0.28
    track
    -0.28
    REQUIRED
    -0.28
     REQUIRED
    -0.28
    Required
    -0.27
    stick
    -0.27
     necessary
    -0.27
     just
    -0.26
    anko
    -0.26
    POSITIVE LOGITS
     right
    0.60
    right
    0.58
    åı³
    0.54
     Right
    0.52
    RIGHT
    0.47
    	right
    0.44
    (right
    0.43
    _right
    0.43
    .right
    0.42
     RIGHT
    0.42
    Activations Density 0.002%

    No Known Activations