© 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. 29088
    Prev
    Next
    INDEX
    Explanations

    conjunctions and punctuation marks indicating a transition or continuation of thought

    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
    èm
    -0.13
    عداد
    -0.13
    GMEM
    -0.13
    meni
    -0.13
     ones
    -0.13
     iParam
    -0.12
    pread
    -0.12
    orias
    -0.12
    Ãłn
    -0.12
    jÃŃm
    -0.12
    POSITIVE LOGITS
     their
    0.86
    Their
    0.76
     Their
    0.75
    their
    0.73
     theirs
    0.66
     THEIR
    0.64
     loro
    0.63
     иÑħ
    0.62
     they
    0.59
     leur
    0.57
    Activations Density 0.035%

    No Known Activations