© 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. 95875
    Prev
    Next
    INDEX
    Explanations

    reflect

    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
    .NewRequest
    -0.28
    atoria
    -0.26
    пон
    -0.25
    /request
    -0.25
     Uniform
    -0.25
    /bower
    -0.24
    å¾®è§Ĥ
    -0.24
    unn
    -0.24
    åľºé¦Ĩ
    -0.24
     Request
    -0.24
    POSITIVE LOGITS
     reflection
    0.55
     reflect
    0.52
     reflected
    0.52
    reflect
    0.50
     reflecting
    0.49
     reflections
    0.48
     reflective
    0.47
     Reflect
    0.46
    Reflect
    0.46
     Reflection
    0.44
    Activations Density 0.894%

    No Known Activations