© 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. 28711
    Prev
    Next
    INDEX
    Explanations

    present tense forms of the verb "to be" along with expressions of surprise or awakening

    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
    ipay
    -0.17
    istrovstvÃŃ
    -0.16
    arkan
    -0.15
     pressure
    -0.15
    astes
    -0.14
    ccak
    -0.14
    thur
    -0.14
    oho
    -0.14
    atik
    -0.14
    xee
    -0.14
    POSITIVE LOGITS
     overcome
    0.33
     struck
    0.30
     hit
    0.28
     filled
    0.24
    hit
    0.24
     envelop
    0.22
     strike
    0.22
     bomb
    0.22
     suddenly
    0.22
    filled
    0.22
    Activations Density 0.038%

    No Known Activations