© 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. 12-TRANSCODER-HP
    7. 23076
    Prev
    Next
    INDEX
    Explanations

    Organization and sorting

    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
    Not in Any Lists

    No Comments

    Negative Logits
    é²ľ
    -0.28
     explains
    -0.27
    解éĩĬ
    -0.27
     Olsen
    -0.26
     δ
    -0.26
    ongoose
    -0.25
     milano
    -0.25
    unable
    -0.24
    åı¯çĪ±çļĦ
    -0.23
    wide
    -0.23
    POSITIVE LOGITS
    æľį
    0.26
    á»ĩnh
    0.25
    éĻ¢éķ¿
    0.25
    Pres
    0.25
     pres
    0.25
    (inst
    0.24
    åı£æľį
    0.24
    preserve
    0.24
    .inst
    0.24
     Helm
    0.24
    Activations Density 0.002%

    No Known Activations