© 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. 107342
    Prev
    Next
    INDEX
    Explanations

    Academic/scientific texts

    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
    kt
    -0.28
     proper
    -0.27
    ivent
    -0.26
     pat
    -0.25
    港
    -0.25
    累累
    -0.24
    ÑĮÑĤе
    -0.24
    åŁł
    -0.24
     Fuji
    -0.24
    ##_
    -0.24
    POSITIVE LOGITS
     senses
    0.28
     cheating
    0.24
    ĵį
    0.24
     forks
    0.24
    è°¦
    0.23
    closure
    0.23
    vm
    0.23
    å¹½
    0.23
    æļ®
    0.23
    -base
    0.23
    Activations Density 11.313%

    No Known Activations