© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 24138
    Prev
    Next
    INDEX
    Explanations

    concepts and phrases related to foundational elements and their importance

    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
     others
    -0.15
    .stub
    -0.15
     Others
    -0.13
     anderen
    -0.13
    namen
    -0.13
    æī¶
    -0.13
    à¸ķรà¸ĩ
    -0.13
    å¨
    -0.13
     mesmo
    -0.12
    INA
    -0.12
    POSITIVE LOGITS
     any
    0.43
     successful
    0.32
     every
    0.30
     success
    0.26
     лÑİбой
    0.26
    ä»»ä½ķ
    0.26
     modern
    0.26
     Successful
    0.25
     most
    0.25
     many
    0.25
    Activations Density 0.259%

    No Known Activations