© 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. 23-TRANSCODER-HP
    7. 24982
    Prev
    Next
    INDEX
    Explanations

    reduce

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    styl
    -0.28
    ç»ĵ
    -0.26
    交æĺĵå¹³åı°
    -0.26
    bat
    -0.26
     infinitely
    -0.24
    æĹıèĩªæ²»å·ŀ
    -0.24
    触åıij
    -0.24
     JFactory
    -0.23
     Decoder
    -0.23
    ustralian
    -0.23
    POSITIVE LOGITS
    ive
    0.33
    arde
    0.30
     redu
    0.28
     ÑģлÑĥÑĩа
    0.28
    ings
    0.28
     ghost
    0.27
     am
    0.26
     guarding
    0.26
     oxidative
    0.26
     pred
    0.24
    Activations Density 0.090%

    No Known Activations