© 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. 9078
    Prev
    Next
    INDEX
    Explanations

    references to the challenges and dynamics of law school graduates in the job market

    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
    .sponge
    -0.16
    aje
    -0.15
    éĢł
    -0.15
    èĩ´
    -0.14
    inear
    -0.14
     Laud
    -0.14
    .Dto
    -0.14
    SEQU
    -0.14
     lamin
    -0.13
    IEW
    -0.13
    POSITIVE LOGITS
     law
    0.58
     legal
    0.54
     Legal
    0.54
     Law
    0.51
    law
    0.48
    Legal
    0.47
     lawyer
    0.45
    -law
    0.45
     lawy
    0.45
    Law
    0.45
    Activations Density 0.048%

    No Known Activations