© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 4-TRANSCODER-HP
    7. 108
    Prev
    Next
    INDEX
    Explanations

    ventional/ational

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_4.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.4.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Nan
    -0.29
     “
    -0.26
     Nah
    -0.26
    èħĶ
    -0.26
    antan
    -0.26
     mâ
    -0.26
     id
    -0.25
     re
    -0.24
     Stream
    -0.23
    罨
    -0.23
    POSITIVE LOGITS
    depend
    0.27
    iesel
    0.27
     Vince
    0.26
    åĬ¨æijĩ
    0.25
    pill
    0.25
    sein
    0.24
    orks
    0.24
    elsen
    0.24
    ixin
    0.24
    werk
    0.24
    Activations Density 0.008%

    No Known Activations