© 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. 32996
    Prev
    Next
    INDEX
    Explanations

    references to news and updates about current events

    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
    assin
    -0.15
    ieri
    -0.15
    phinx
    -0.15
    mented
    -0.14
    _ISR
    -0.14
     Taken
    -0.14
    assage
    -0.13
    andering
    -0.13
     Escort
    -0.13
    ãĥ³ãĥķ
    -0.13
    POSITIVE LOGITS
     spoke
    0.29
     contact
    0.26
     contacted
    0.26
     contacting
    0.23
     obtained
    0.23
    contact
    0.22
     spoken
    0.22
     attempted
    0.21
     unable
    0.21
     attempts
    0.20
    Activations Density 0.058%

    No Known Activations