© 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. 29608
    Prev
    Next
    INDEX
    Explanations

    discussions concerning race and gender dynamics in professional contexts

    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
    leurs
    -0.15
    -urlencoded
    -0.15
    isia
    -0.14
    ollar
    -0.14
    avar
    -0.14
    alers
    -0.14
    olvers
    -0.13
    éĺħ读次æķ°
    -0.13
    ÑĢава
    -0.13
    aldi
    -0.13
    POSITIVE LOGITS
     unnamed
    0.23
     representative
    0.23
     anonymous
    0.21
     guy
    0.19
     anon
    0.19
     witness
    0.18
     unidentified
    0.18
     interviewer
    0.18
     caller
    0.18
     colleague
    0.18
    Activations Density 0.187%

    No Known Activations