© 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. 35-TRANSCODER-HP
    7. 40360
    Prev
    Next
    INDEX
    Explanations

    technical reports

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    鸳
    -0.29
     Rod
    -0.27
    kos
    -0.27
    æĪijä¸į
    -0.26
     Rita
    -0.26
     RDD
    -0.25
    壬
    -0.25
    ëĵľ
    -0.24
    使çĶ¨çļĦ
    -0.24
    æĿĥ
    -0.24
    POSITIVE LOGITS
    develop
    0.30
    incer
    0.30
    ception
    0.30
    establish
    0.29
    ierarchy
    0.29
    trusted
    0.29
    commit
    0.29
    iphery
    0.28
    andles
    0.28
    legation
    0.28
    Activations Density 2.069%

    No Known Activations