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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 28
    Prev
    Next
    INDEX
    Explanations

    expressions of apology or regret

    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
     (
    -0.17
    ,
    -0.17
    ulner
    -0.15
     tinh
    -0.15
     Jeffrey
    -0.14
    :
    -0.14
    è«
    -0.14
    ni
    -0.14
     online
    -0.14
     universal
    -0.14
    POSITIVE LOGITS
     lengthy
    0.23
    ยาว
    0.21
     dlouh
    0.20
     LENGTH
    0.20
     overposting
    0.19
     Ø·ÙĪÙĦ
    0.19
     dÃłi
    0.19
    -long
    0.19
    LONG
    0.19
     length
    0.18
    Activations Density 0.057%

    No Known Activations