© 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. 14-TRANSCODER-HP
    7. 40179
    Prev
    Next
    INDEX
    Explanations

    punctuation and filler words

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.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¿ĥçģµ
    -0.28
    æ·º
    -0.26
     struggled
    -0.26
    ilit
    -0.26
    管çIJĨ
    -0.26
    æĹłåģ¿
    -0.26
     manages
    -0.25
     composed
    -0.25
    huge
    -0.25
     manage
    -0.24
    POSITIVE LOGITS
    بت
    0.26
    æĵ¦
    0.26
    çĹĺ
    0.24
    å±ĭ
    0.24
    _Enter
    0.24
    æīĴ
    0.23
     Hawks
    0.23
    ires
    0.23
    ÑĢог
    0.23
    `.`
    0.23
    Activations Density 0.896%

    No Known Activations