Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 15869
    Prev
    Next
    INDEX
    Explanations

    elements related to competitive racing scenarios

    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
    á»Ļng
    -0.15
    ewire
    -0.14
    zdy
    -0.14
    èĢ
    -0.14
     satellite
    -0.14
     maduras
    -0.13
    fone
    -0.13
     Laud
    -0.13
    è¨Ģ
    -0.13
     Comm
    -0.13
    POSITIVE LOGITS
     Turns
    0.32
     turns
    0.32
     turn
    0.32
     curve
    0.28
     Turn
    0.28
    Turn
    0.28
    turn
    0.27
     corners
    0.26
     curves
    0.26
     bend
    0.26
    Activations Density 0.011%

    No Known Activations