Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 27-TRANSCODER-HP
    4. 1171
    Prev
    Next
    INDEX
    Explanations

    code or articles titles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŀ¶
    -0.27
    antt
    -0.27
    undle
    -0.26
    uble
    -0.25
     exercised
    -0.24
     exercising
    -0.24
     strokes
    -0.24
     Esc
    -0.24
    ¹Ħ
    -0.24
    oop
    -0.24
    POSITIVE LOGITS
     LS
    0.29
    heits
    0.26
    cel
    0.25
    å®ļåIJij
    0.25
    SHOT
    0.24
    thesis
    0.24
    LS
    0.24
    ëį°
    0.23
    Tex
    0.23
     ls
    0.23
    Activations Density 0.025%

    No Known Activations