© 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. 63460
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the initiation of a dialogue or event

    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
     ConnectionState
    -0.15
    anzi
    -0.14
    Completion
    -0.14
    eland
    -0.14
    reste
    -0.14
    Decision
    -0.14
    wick
    -0.14
    Chance
    -0.14
    nds
    -0.13
    .sig
    -0.13
    POSITIVE LOGITS
     journey
    0.28
     Engines
    0.24
     engines
    0.23
     new
    0.23
     anew
    0.23
     process
    0.23
     countdown
    0.22
     conversation
    0.21
    jour
    0.21
    æĸ°çļĦ
    0.20
    Activations Density 0.031%

    No Known Activations