© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 15797
    Prev
    Next
    INDEX
    Explanations

    instances of the word "it" and related expressions indicating consequence or hypothetical scenarios

    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
    kyt
    -0.17
    edith
    -0.15
    ulist
    -0.15
    ãĥĶãĥ¼
    -0.15
    enance
    -0.14
    .onStart
    -0.14
    ulumi
    -0.14
    erin
    -0.14
    iants
    -0.14
    deny
    -0.14
    POSITIVE LOGITS
     because
    0.21
     becomes
    0.21
     mean
    0.20
     means
    0.20
     beh
    0.19
    mean
    0.19
    iner
    0.19
     will
    0.18
    'll
    0.18
     Mean
    0.18
    Activations Density 0.065%

    No Known Activations