© 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. 41265
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate differing opinions or perspectives on topics

    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
    (æĹ¥
    -0.15
     anderen
    -0.14
    .wik
    -0.13
    ä¸įåŃĺåľ¨
    -0.13
    adil
    -0.13
     вÑĸдповÑĸд
    -0.12
    ÅĻiv
    -0.12
    Ú©ÛĮÙĦ
    -0.12
     keinen
    -0.12
    others
    -0.12
    POSITIVE LOGITS
     this
    0.51
    this
    0.39
     these
    0.39
    	this
    0.30
    these
    0.28
    (this
    0.27
    è¿Ļ
    0.27
     THIS
    0.27
     it
    0.26
    ãģĵãģ®
    0.25
    Activations Density 0.580%

    No Known Activations