© 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. 63280
    Prev
    Next
    INDEX
    Explanations

    references to the act of turning or changing direction, particularly in the context of policy or behavior

    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
    otes
    -0.16
    anou
    -0.15
    207
    -0.15
    strup
    -0.15
    iola
    -0.14
    letic
    -0.14
    κοÏį
    -0.14
    енка
    -0.14
    ãĥ¼ãĥĪ
    -0.14
    iser
    -0.13
    POSITIVE LOGITS
     Turn
    0.28
     screws
    0.26
     dial
    0.26
     turn
    0.26
    -turn
    0.25
    (turn
    0.24
     turning
    0.24
     corner
    0.24
     tide
    0.24
    Turn
    0.24
    Activations Density 0.017%

    No Known Activations