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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 8661
    Prev
    Next
    INDEX
    Explanations

    phrases related to educational and career aspirations

    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
    enia
    -0.17
    aval
    -0.15
     assign
    -0.15
    ovel
    -0.15
    Thumb
    -0.15
    aju
    -0.14
    ANTE
    -0.13
    arov
    -0.13
    imoto
    -0.13
    icher
    -0.13
    POSITIVE LOGITS
     pursue
    0.39
     pursuing
    0.38
     purs
    0.36
     pursued
    0.35
     Purs
    0.33
     pursuit
    0.32
     pursuits
    0.30
     studying
    0.28
     careers
    0.27
     becoming
    0.26
    Activations Density 0.051%

    No Known Activations