© 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. 29011
    Prev
    Next
    INDEX
    Explanations

    conditional phrases starting with "if"

    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
    ovky
    -0.15
    ردÙĩ
    -0.14
    APON
    -0.13
     адже
    -0.13
    iets
    -0.13
    нав
    -0.13
    Associated
    -0.13
    IGN
    -0.13
    ì§Ŀ
    -0.12
    .sys
    -0.12
    POSITIVE LOGITS
     things
    0.31
     nothing
    0.30
     indeed
    0.29
     anything
    0.28
     everything
    0.25
     somehow
    0.25
     anyone
    0.24
    /
    0.24
     they
    0.23
    things
    0.23
    Activations Density 0.138%

    No Known Activations