© 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. 29516
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names of people and organizations

    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
    questions
    -0.15
     serialVersionUID
    -0.15
     question
    -0.14
    åķı
    -0.14
    Ask
    -0.14
    ask
    -0.14
     Questions
    -0.13
    çĶ
    -0.13
    문ìĿĺ
    -0.13
    ุà¹Ī
    -0.13
    POSITIVE LOGITS
     added
    0.38
     said
    0.32
    added
    0.26
    Added
    0.24
    -added
    0.23
     Added
    0.23
     adds
    0.22
     adding
    0.20
    _added
    0.20
     add
    0.20
    Activations Density 0.046%

    No Known Activations