© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41705
    Prev
    Next
    INDEX
    Explanations

    references to rights and social justice issues

    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
    odesk
    -0.15
    ulator
    -0.15
    rame
    -0.15
     stretch
    -0.14
    kees
    -0.14
    iker
    -0.14
    eldorf
    -0.14
    mÄĽ
    -0.14
    ä¸į足
    -0.14
     skew
    -0.14
    POSITIVE LOGITS
     regression
    0.50
     regress
    0.49
     backward
    0.47
     retro
    0.45
     backwards
    0.45
     Regression
    0.41
     reg
    0.39
     Retro
    0.39
    Regression
    0.39
    åĽŀ
    0.37
    Activations Density 0.030%

    No Known Activations