© 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. 35-TRANSCODER-HP
    7. 40423
    Prev
    Next
    INDEX
    Explanations

    Code and website text

    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
    scape
    -0.29
    ç°§
    -0.28
    losures
    -0.28
    odore
    -0.27
    bor
    -0.26
    shire
    -0.25
    è´¬
    -0.25
    ificates
    -0.25
    neath
    -0.25
    arms
    -0.24
    POSITIVE LOGITS
    éļ¾çĤ¹
    0.26
    isa
    0.26
    Difficulty
    0.25
    æ·±å¤Ħ
    0.25
    åĬłåħ¥
    0.24
    å¥ij
    0.24
    iso
    0.24
    缴å±ŀ
    0.24
    常åĬ¡
    0.24
    uria
    0.24
    Activations Density 0.001%

    No Known Activations