© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 24156
    Prev
    Next
    INDEX
    Explanations

    statements and promises made within conversations

    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
     granddaughter
    -0.18
     friend
    -0.17
     grandson
    -0.17
    MF
    -0.16
     colleague
    -0.15
    æľĭåıĭ
    -0.15
    udos
    -0.15
     niece
    -0.15
     grandchildren
    -0.14
     nephew
    -0.14
    POSITIVE LOGITS
     Mom
    0.57
     mom
    0.53
    Mom
    0.52
     Dad
    0.46
    mom
    0.45
     dad
    0.44
     MOM
    0.42
    dad
    0.39
     mommy
    0.36
     Mum
    0.33
    Activations Density 0.027%

    No Known Activations