© 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. 29509
    Prev
    Next
    INDEX
    Explanations

    phrases indicating reflection and questioning of statements made by individuals

    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
    peria
    -0.17
    åŃĺäºİ
    -0.16
    iros
    -0.16
    regunta
    -0.16
    LOAT
    -0.16
    mak
    -0.15
    REFERRED
    -0.15
    dül
    -0.14
    lut
    -0.14
     McKay
    -0.14
    POSITIVE LOGITS
     said
    0.41
     say
    0.39
     saying
    0.37
     says
    0.35
     Say
    0.30
     Said
    0.28
    Say
    0.28
    说
    0.27
    say
    0.26
     SAY
    0.26
    Activations Density 0.065%

    No Known Activations