© 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. 23917
    Prev
    Next
    INDEX
    Explanations

    references to government projects and financing details

    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
     spoken
    -0.20
     grown
    -0.18
     flown
    -0.18
     beaten
    -0.18
    dden
    -0.17
     lain
    -0.16
     bitten
    -0.16
     eaten
    -0.16
     shaken
    -0.16
    oken
    -0.16
    POSITIVE LOGITS
    rió
    0.33
     hizo
    0.30
     yaptı
    0.30
    ÏĦηÏĥε
    0.29
    uvo
    0.29
    ÏīÏĥε
    0.27
     aldı
    0.27
    äºĨä¸Ģ
    0.27
     gave
    0.27
    ió
    0.27
    Activations Density 0.016%

    No Known Activations