© 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. 33-TRANSCODER-HP
    7. 54128
    Prev
    Next
    INDEX
    Explanations

    software license information

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å±ħ
    -0.32
    éħįå¥Ĺ
    -0.30
    ittest
    -0.29
    елÑĮ
    -0.29
    æī¶
    -0.28
     forma
    -0.25
    æİĺ
    -0.25
     univers
    -0.25
     Plans
    -0.24
     civil
    -0.23
    POSITIVE LOGITS
    following
    0.37
     Following
    0.35
     following
    0.34
    Following
    0.29
    ä¸ĭæĿ¥çļĦ
    0.28
    å¾Ģä¸ĭ
    0.26
     Masks
    0.25
    æİĤ
    0.25
     seguint
    0.25
    ä¸įå±ij
    0.25
    Activations Density 0.007%

    No Known Activations