© 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. 22-TRANSCODER-HP
    7. 112422
    Prev
    Next
    INDEX
    Explanations

    code/non-natural language

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĹ¶ä»£
    -0.35
    -dem
    -0.33
    æĹ¶ä»£çļĦ
    -0.32
     demands
    -0.31
    çļĦæĹ¶ä»£
    -0.30
    çļĦè¦ģæ±Ĥ
    -0.29
     exig
    -0.28
    æĻĤ代
    -0.27
    è¦ģæ±Ĥ
    -0.26
    æ¼Ķåıĺ
    -0.25
    POSITIVE LOGITS
    apot
    0.26
     Segment
    0.25
     panic
    0.25
    ability
    0.25
    udos
    0.24
    åijµæĬ¤
    0.24
    ensual
    0.24
     buffering
    0.24
    IFO
    0.23
     ROM
    0.23
    Activations Density 0.000%

    No Known Activations