© 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. 15643
    Prev
    Next
    INDEX
    Explanations

    references to companies and their actions or involvement in specific situations

    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
     itself
    -0.22
     its
    -0.17
     Its
    -0.16
    Its
    -0.16
    ستاÙĨ
    -0.15
     Alone
    -0.15
    æĺ¯ä¸Ģ个
    -0.14
    ersistence
    -0.14
    uka
    -0.14
    _PUSHDATA
    -0.14
    POSITIVE LOGITS
     themselves
    0.48
     their
    0.33
    Their
    0.29
     Their
    0.29
     increasingly
    0.26
    their
    0.26
     alike
    0.23
     leurs
    0.23
     loro
    0.22
     nhau
    0.21
    Activations Density 0.103%

    No Known Activations