© 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. 141488
    Prev
    Next
    INDEX
    Explanations

    harm

    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
    涵çĽĸäºĨ
    -0.30
    çĻ½èī²çļĦ
    -0.27
    åĵī
    -0.27
    æĹĥ
    -0.27
    ä½ļ
    -0.27
    åĮħåIJ«äºĨ
    -0.26
    åľ°å¸¦
    -0.26
    å¿ĺè®°äºĨ
    -0.25
    -present
    -0.25
    çĶµéĺ»
    -0.25
    POSITIVE LOGITS
     propri
    0.30
     pol
    0.27
    vol
    0.26
    é¾Ħ
    0.26
     Int
    0.25
    ground
    0.25
     Vol
    0.25
     raw
    0.25
    çģµ
    0.25
    shake
    0.25
    Activations Density 5.866%

    No Known Activations