© 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. 41057
    Prev
    Next
    INDEX
    Explanations

    references to a group of people or entities, particularly using the word "those."

    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
    ÅĦ
    -0.14
    dık
    -0.14
    odal
    -0.13
    ude
    -0.13
    äºŃ
    -0.13
    hang
    -0.13
     outset
    -0.13
    agram
    -0.12
    .Elapsed
    -0.12
    illing
    -0.12
    POSITIVE LOGITS
     are
    0.57
     were
    0.42
     aren
    0.42
     weren
    0.34
     são
    0.33
     Are
    0.32
     ARE
    0.31
     sont
    0.30
    _are
    0.29
     jsou
    0.28
    Activations Density 0.066%

    No Known Activations