© 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. 107137
    Prev
    Next
    INDEX
    Explanations

    Technical/statistical contexts

    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
    clock
    -0.28
    ody
    -0.27
    ODY
    -0.26
     clock
    -0.26
    zeich
    -0.25
    ((&
    -0.25
    缮çĿ¹
    -0.25
    ously
    -0.25
    Pri
    -0.24
     Gould
    -0.24
    POSITIVE LOGITS
    æĺ¯ä»¥
    0.32
    which
    0.29
    =Integer
    0.28
    HOOK
    0.28
     which
    0.26
     vá»ijn
    0.26
    ignite
    0.25
     (=
    0.24
     CAST
    0.24
     (âĪ
    0.24
    Activations Density 44.302%

    No Known Activations