© 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. 18-TRANSCODER-HP
    7. 88608
    Prev
    Next
    INDEX
    Explanations

    code punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     rehears
    -0.28
    +-+-
    -0.28
    osaic
    -0.28
    _simps
    -0.25
    åĥ¦
    -0.25
    	sys
    -0.25
    ÑĤика
    -0.24
    åİ¢
    -0.24
    reuse
    -0.23
    .subscribe
    -0.23
    POSITIVE LOGITS
    以èī²
    0.28
    ettel
    0.27
    olang
    0.25
    Trait
    0.25
     parameter
    0.25
    intern
    0.24
    overs
    0.24
    sn
    0.24
    éĥ¤
    0.24
     лиÑĪ
    0.24
    Activations Density 0.024%

    No Known Activations