© 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. 22-TRANSCODER-HP
    7. 98095
    Prev
    Next
    INDEX
    Explanations

    conditional situations, potential problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åīįãģ®
    -0.27
    whole
    -0.26
     getSource
    -0.26
    âĪ¬
    -0.26
     ICommand
    -0.25
    æĿ¥åİĨ
    -0.25
    =all
    -0.24
    history
    -0.24
    ativity
    -0.24
    ilar
    -0.24
    POSITIVE LOGITS
     later
    0.40
    失åĪ©
    0.39
     Later
    0.38
    失误
    0.37
    失败
    0.36
    Later
    0.35
     faulty
    0.33
    æķĹ
    0.32
     defeat
    0.31
    è´¥
    0.31
    Activations Density 0.007%

    No Known Activations