© 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. 20-TRANSCODER-HP
    7. 141677
    Prev
    Next
    INDEX
    Explanations

    Scientific species and genes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĶ¾å¤§
    -0.27
    该æĢİä¹ĪåĬŀ
    -0.26
    uple
    -0.25
     enlargement
    -0.25
     tac
    -0.25
    -done
    -0.25
    thon
    -0.25
     elevate
    -0.24
    大夫
    -0.24
     spre
    -0.24
    POSITIVE LOGITS
    der
    0.34
    nder
    0.29
    ABL
    0.26
    taire
    0.26
    *I
    0.26
    trs
    0.25
    è¡į
    0.25
    HF
    0.25
    DER
    0.24
    ared
    0.24
    Activations Density 0.618%

    No Known Activations