Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 31362
    Prev
    Next
    INDEX
    Explanations

    participles and forms of the verb "to be"

    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
    inç
    -0.17
    -lfs
    -0.16
     NavParams
    -0.15
    bild
    -0.15
    Äįel
    -0.14
    ãĥĥãĥĦ
    -0.14
    agi
    -0.14
    ãģªãģĮ
    -0.14
    icies
    -0.14
    OnUiThread
    -0.14
    POSITIVE LOGITS
     away
    0.31
    -away
    0.26
     Away
    0.25
    Away
    0.24
    -out
    0.23
     out
    0.22
     off
    0.21
     into
    0.21
     onto
    0.20
    -up
    0.20
    Activations Density 0.066%

    No Known Activations