© 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. 29154
    Prev
    Next
    INDEX
    Explanations

    personal pronouns and phrases indicating subjective opinions or assertions

    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
     note
    -0.17
     says
    -0.15
     WARRANT
    -0.15
     that
    -0.15
     seh
    -0.15
     bekannt
    -0.15
     ensure
    -0.14
     clearly
    -0.14
     tells
    -0.14
     kidding
    -0.14
    POSITIVE LOGITS
     indeed
    0.25
    езÑĥлÑĮÑĤ
    0.16
    etsk
    0.16
    ưỡng
    0.15
    ÃŃo
    0.15
     Indeed
    0.15
     aslında
    0.14
     realmente
    0.14
     konkrét
    0.14
    aming
    0.14
    Activations Density 0.044%

    No Known Activations