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. 23742
    Prev
    Next
    INDEX
    Explanations

    phrases related to collaboration and bringing people or ideas together

    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
    /from
    -0.19
    Ïĥί
    -0.16
    лини
    -0.15
    acea
    -0.14
    akistan
    -0.14
    /on
    -0.13
    bras
    -0.13
    ÑĨен
    -0.13
    ysa
    -0.13
    -runtime
    -0.13
    POSITIVE LOGITS
     forth
    0.45
     together
    0.33
     down
    0.27
     back
    0.27
    ToFront
    0.27
     about
    0.25
    forth
    0.24
     attention
    0.23
     balance
    0.21
    down
    0.21
    Activations Density 0.043%

    No Known Activations