© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 32895
    Prev
    Next
    INDEX
    Explanations

    references to WWE programming and its evaluation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rese
    -0.16
    cad
    -0.15
    itech
    -0.14
    abela
    -0.14
    roof
    -0.14
    Mapper
    -0.14
    leton
    -0.13
     Newsp
    -0.13
    鸡
    -0.13
    western
    -0.13
    POSITIVE LOGITS
     Sm
    0.26
     Raw
    0.26
     RAW
    0.24
     developmental
    0.24
     WWE
    0.22
    Raw
    0.20
    RAW
    0.20
     NXT
    0.20
     kay
    0.19
     storyline
    0.19
    Activations Density 0.018%

    No Known Activations