© 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. 30-TRANSCODER-HP
    7. 96178
    Prev
    Next
    INDEX
    Explanations

    Code/technical documentation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     past
    -0.27
     Gladiator
    -0.26
    Past
    -0.26
    磾
    -0.25
     Past
    -0.25
    éļİ
    -0.25
    éĺ¶
    -0.24
    Folders
    -0.24
    past
    -0.24
    âĨ©
    -0.23
    POSITIVE LOGITS
    abinet
    0.30
    å¼ĢæĶ¾å¼ı
    0.29
    felt
    0.28
    ByExample
    0.26
     suicide
    0.26
    æĬ±çĿĢ
    0.26
    \Schema
    0.25
    subscribe
    0.25
    _<?
    0.24
     corpor
    0.24
    Activations Density 0.001%

    No Known Activations