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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 48
    Prev
    Next
    INDEX
    Explanations

    phrases expressing conditional statements or suggestions

    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
    ovic
    -0.20
    ↵↵
    -0.17
    ÙĨاÙĨ
    -0.15
    elize
    -0.15
    sov
    -0.15
    itom
    -0.15
    ovich
    -0.14
    окол
    -0.14
    itr
    -0.14
    änger
    -0.14
    POSITIVE LOGITS
     focus
    0.22
     instead
    0.20
     stop
    0.20
    Focus
    0.20
     Facts
    0.19
    facts
    0.19
     facts
    0.19
    focus
    0.19
     reality
    0.18
     Focus
    0.18
    Activations Density 0.075%

    No Known Activations