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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 25122
    Prev
    Next
    INDEX
    Explanations

    references to surveys and estimation processes related to job and economic data

    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
    ayout
    -0.15
    @Before
    -0.14
    ogan
    -0.14
    oje
    -0.13
    aucoup
    -0.13
     ActionTypes
    -0.13
    agy
    -0.13
    rai
    -0.13
    InnerHTML
    -0.13
     preced
    -0.13
    POSITIVE LOGITS
     starting
    1.07
     beginning
    0.96
     Starting
    0.90
    starting
    0.87
    Starting
    0.85
     Beginning
    0.81
    Beginning
    0.73
    begin
    0.60
     effective
    0.46
     begin
    0.46
    Activations Density 0.033%

    No Known Activations