© 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. 41260
    Prev
    Next
    INDEX
    Explanations

    phrases related to creation and aggregation of information or items

    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
    ammad
    -0.15
    .TestTools
    -0.15
    amar
    -0.14
    .si
    -0.14
    itchens
    -0.14
    riter
    -0.14
    à¹Ģà¸Ĺà¸ŀ
    -0.13
     seins
    -0.13
    aven
    -0.13
    .omg
    -0.12
    POSITIVE LOGITS
     create
    0.59
     creates
    0.55
     creating
    0.55
     forming
    0.52
    .create
    0.46
    create
    0.45
     Create
    0.45
    å½¢æĪIJ
    0.43
    creating
    0.43
     produce
    0.43
    Activations Density 0.057%

    No Known Activations