© 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. 30063
    Prev
    Next
    INDEX
    Explanations

    references to events or actions in the context of a group or collective experience

    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
     Seconds
    -0.17
    rush
    -0.17
    jets
    -0.15
    ãĥ³ãĥij
    -0.14
     Flutter
    -0.14
     Door
    -0.14
    gow
    -0.14
     door
    -0.13
    zap
    -0.13
    subcategory
    -0.13
    POSITIVE LOGITS
     camp
    0.27
     camps
    0.22
     crosses
    0.20
     bush
    0.19
     skirt
    0.18
     Camp
    0.17
    260
    0.17
     split
    0.17
    camp
    0.17
     crossed
    0.17
    Activations Density 0.079%

    No Known Activations