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. 23635
    Prev
    Next
    INDEX
    Explanations

    discussions about the impact of careers and personal stakes in relation to decision-making and beliefs

    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
    icut
    -0.19
    iel
    -0.15
    ardo
    -0.15
    ifo
    -0.14
    à¸Ľà¸ı
    -0.14
     wherever
    -0.14
    ocl
    -0.14
    nage
    -0.13
    939
    -0.13
    81
    -0.13
    POSITIVE LOGITS
     stakes
    0.62
     stake
    0.53
     Stake
    0.39
    stakes
    0.38
     hangs
    0.33
     hang
    0.31
    Hang
    0.28
     riding
    0.28
     hanging
    0.28
     rides
    0.28
    Activations Density 0.036%

    No Known Activations