© 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 EvalsExportsAPI 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. 14071
    Prev
    Next
    INDEX
    Explanations

    expressions of personal uncertainty or indecision regarding career paths

    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
     accomplish
    -0.18
     accompl
    -0.17
    fol
    -0.17
     Mom
    -0.16
     accomplishment
    -0.16
    edith
    -0.15
    Mom
    -0.15
     Huffman
    -0.14
     mom
    -0.14
     priv
    -0.14
    POSITIVE LOGITS
     bits
    0.21
     aload
    0.20
     Sorted
    0.19
     sort
    0.19
     lovely
    0.19
    Sorted
    0.18
     erm
    0.18
     mates
    0.18
     proper
    0.18
     sorted
    0.18
    Activations Density 0.292%

    No Known Activations