© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 40823
    Prev
    Next
    INDEX
    Explanations

    repeated phrases or structures indicating collective or coordinated ideas

    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
     @{
    -0.15
    icros
    -0.14
    ocaly
    -0.14
     sanity
    -0.14
    ulet
    -0.14
    869
    -0.13
    adox
    -0.13
    px
    -0.13
     Leadership
    -0.13
     Enemy
    -0.13
    POSITIVE LOGITS
     vice
    0.21
     other
    0.20
    åī¯
    0.18
    ODO
    0.18
     deputy
    0.17
    ients
    0.16
     others
    0.16
    asje
    0.15
     assistant
    0.15
    ungan
    0.15
    Activations Density 0.038%

    No Known Activations