© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 41119
    Prev
    Next
    INDEX
    Explanations

    interactions involving female speakers

    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
    ubat
    -0.16
    .intellij
    -0.16
    untime
    -0.16
    ÑģÑıг
    -0.15
    osg
    -0.15
    strcasecmp
    -0.14
    çģ
    -0.14
    RecognitionException
    -0.14
    argon
    -0.14
     нÑĸби
    -0.14
    POSITIVE LOGITS
     reply
    0.51
     replied
    0.48
     replies
    0.46
     Reply
    0.41
    reply
    0.40
     answer
    0.39
     response
    0.39
     responded
    0.36
    Reply
    0.36
     Answer
    0.36
    Activations Density 0.056%

    No Known Activations