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. 23662
    Prev
    Next
    INDEX
    Explanations

    first-person pronouns and expressions of uncertainty or speculation

    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
    ãĥ«ãĥķ
    -0.14
    adx
    -0.14
    acey
    -0.14
    @show
    -0.14
    524
    -0.14
     èµ·
    -0.13
    hood
    -0.13
    ies
    -0.13
    åŀ
    -0.13
    IVED
    -0.13
    POSITIVE LOGITS
     don
    0.60
    don
    0.50
     Don
    0.44
     doesn
    0.44
     dont
    0.43
    Don
    0.43
     DON
    0.42
    ä¸įçŁ¥éģĵ
    0.39
     Dun
    0.36
     dun
    0.36
    Activations Density 0.045%

    No Known Activations