© 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. 95455
    Prev
    Next
    INDEX
    Explanations

    is

    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
    æŶéĹ´åĴĮ
    -0.29
    ught
    -0.28
     вÑĢемени
    -0.27
     specialized
    -0.27
     found
    -0.27
     TIME
    -0.26
    æĻĤéĸĵ
    -0.25
     time
    -0.25
    ainted
    -0.25
    ski
    -0.24
    POSITIVE LOGITS
     when
    0.37
    çªĸ
    0.31
    å®ĥ
    0.28
    å½ĵ
    0.28
     When
    0.27
    ãĥ¼ãĥł
    0.27
     khi
    0.27
     it
    0.27
    When
    0.27
    å½ĵä»ĸ
    0.26
    Activations Density 0.789%

    No Known Activations