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. 23164
    Prev
    Next
    INDEX
    Explanations

    indications of commitment and resource allocation towards a project or goal

    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
     their
    -0.24
     themselves
    -0.24
     today
    -0.21
    Their
    -0.21
     Their
    -0.21
    their
    -0.19
     THEIR
    -0.19
     they
    -0.19
     ours
    -0.18
    today
    -0.17
    POSITIVE LOGITS
     myself
    0.53
     my
    0.39
    æĺ¯æĪij
    0.34
    æĪijçļĦ
    0.33
     mijn
    0.31
     saya
    0.30
     minha
    0.29
     meu
    0.28
    my
    0.28
     meinem
    0.27
    Activations Density 0.204%

    No Known Activations