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. 23740
    Prev
    Next
    INDEX
    Explanations

    references to assignments and systems in various contexts, particularly those involving data collection or tasks within a structured framework

    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
    ansom
    -0.14
    ","");↵
    -0.13
    ','');↵
    -0.13
     )↵↵↵↵↵↵↵↵
    -0.13
    adle
    -0.12
    ering
    -0.12
    arel
    -0.12
    âĢIJ
    -0.12
    exus
    -0.12
    ï¼ļ
    -0.12
    POSITIVE LOGITS
     plus
    0.75
    plus
    0.57
    以åıĬ
    0.56
     along
    0.56
     as
    0.52
     PLUS
    0.50
     ainsi
    0.50
    along
    0.49
     sowie
    0.49
     oraz
    0.49
    Activations Density 0.146%

    No Known Activations