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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 31306
    Prev
    Next
    INDEX
    Explanations

    terms related to ongoing economic challenges and conditions

    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
    tu
    -0.17
    xit
    -0.15
       
    -0.15
    ahi
    -0.14
    327
    -0.14
    iro
    -0.14
     Hue
    -0.14
     PureComponent
    -0.14
    junction
    -0.14
     Hv
    -0.13
    POSITIVE LOGITS
    -motion
    0.23
     motion
    0.19
     underst
    0.17
    lers
    0.17
    antal
    0.16
    以æĿ¥
    0.16
    mente
    0.16
    ngrx
    0.16
    ently
    0.16
    groupBox
    0.15
    Activations Density 0.030%

    No Known Activations