© 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. 41191
    Prev
    Next
    INDEX
    Explanations

    instances of victimization and oppression narratives

    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
    .epsilon
    -0.16
    uren
    -0.15
    riz
    -0.15
    ADDE
    -0.15
    erras
    -0.15
    prit
    -0.15
    cli
    -0.14
    ırak
    -0.14
    åªĴ
    -0.14
    asurer
    -0.14
    POSITIVE LOGITS
     victim
    0.36
     Victim
    0.30
     victims
    0.29
    victim
    0.27
     unfairly
    0.26
     entitled
    0.25
     mist
    0.23
     innocent
    0.23
     Victims
    0.23
     targets
    0.23
    Activations Density 0.068%

    No Known Activations