© 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. 39119
    Prev
    Next
    INDEX
    Explanations

    Non-English and code snippets

    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
    iscrim
    -0.31
    icter
    -0.28
    éĻIJ
    -0.26
    éĤ£è¾¹
    -0.25
    irth
    -0.25
    zej
    -0.25
    峡
    -0.25
    artisan
    -0.24
    usted
    -0.24
    olesale
    -0.24
    POSITIVE LOGITS
    ä¸į失
    0.32
    æľŁéĹ´
    0.30
     during
    0.27
    During
    0.27
     Dao
    0.26
     elbow
    0.26
    Duration
    0.26
    benchmark
    0.25
     }//
    0.25
     duration
    0.24
    Activations Density 19.480%

    No Known Activations