© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 29738
    Prev
    Next
    INDEX
    Explanations

    references to collections or anthologies of stories or works

    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
    erdem
    -0.17
    .ibatis
    -0.16
    mentions
    -0.15
    é£
    -0.14
    chedulers
    -0.14
    ãĥ©ãĤ¹
    -0.14
    ziej
    -0.14
    .GetEnumerator
    -0.14
     Cookbook
    -0.14
    chure
    -0.14
    POSITIVE LOGITS
     essays
    0.33
     short
    0.27
     stories
    0.25
     personal
    0.25
     articles
    0.25
     Essays
    0.24
     original
    0.24
     previously
    0.22
     poems
    0.22
    short
    0.22
    Activations Density 0.039%

    No Known Activations