© 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. 29591
    Prev
    Next
    INDEX
    Explanations

    actions and their consequences in various contexts

    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
    oji
    -0.15
    IRA
    -0.15
    roles
    -0.15
    ñas
    -0.14
    ÙĨÙħ
    -0.14
     creampie
    -0.14
    CanBe
    -0.14
    áºŃt
    -0.14
    ÑĢой
    -0.13
    parentNode
    -0.13
    POSITIVE LOGITS
     having
    0.48
    having
    0.41
     had
    0.40
     Having
    0.40
    Having
    0.38
     have
    0.34
     exert
    0.33
     Have
    0.33
    Have
    0.32
    had
    0.32
    Activations Density 0.053%

    No Known Activations