© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 110989
    Prev
    Next
    INDEX
    Explanations

    stories

    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
    Negative Logits
    è§ģ
    -0.32
    æķħ
    -0.27
    满åĪĨ
    -0.26
    让æĤ¨
    -0.26
    ç¿ģ
    -0.26
    以为
    -0.25
    建议
    -0.25
    èİ·å¥ĸ
    -0.25
     surprisingly
    -0.24
    åı¦
    -0.24
    POSITIVE LOGITS
    avour
    0.28
    ffa
    0.27
    ä¸į论æĺ¯
    0.26
    没éĴ±
    0.26
     fulfil
    0.25
    éĺĬ
    0.24
    亢
    0.24
    noinspection
    0.24
    ayment
    0.24
     Proper
    0.24
    Activations Density 0.011%

    No Known Activations