© 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. 12-TRANSCODER-HP
    7. 24196
    Prev
    Next
    INDEX
    Explanations

    codes, abbreviations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åŁ¹åħ»åѦçĶŁ
    -0.29
    esters
    -0.28
    odel
    -0.27
    BUFF
    -0.26
     Intr
    -0.26
    éĥ½è®¤ä¸º
    -0.26
    çĨı
    -0.25
    æĪij认为
    -0.25
     PUBLIC
    -0.24
    estic
    -0.24
    POSITIVE LOGITS
    -taking
    0.28
    .isnan
    0.26
     past
    0.25
    TIM
    0.25
     gears
    0.25
    ừng
    0.24
    infinity
    0.24
     envis
    0.24
    elle
    0.24
    unbind
    0.23
    Activations Density 0.003%

    No Known Activations