© 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. 30157
    Prev
    Next
    INDEX
    Explanations

    phrases related to loss and gratitude in challenging times

    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
     Authority
    -0.16
     authority
    -0.14
    izu
    -0.14
    abor
    -0.14
    arsers
    -0.14
    ugo
    -0.13
    PLICATE
    -0.13
    TYPES
    -0.13
    IBILITY
    -0.13
     lucrative
    -0.13
    POSITIVE LOGITS
     wonderful
    0.41
     lovely
    0.29
     amazing
    0.25
     wonderfully
    0.25
     dear
    0.25
     terrific
    0.25
     dedicated
    0.24
     magnificent
    0.24
     incredible
    0.24
     fantastic
    0.24
    Activations Density 0.056%

    No Known Activations