© 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. 54728
    Prev
    Next
    INDEX
    Explanations

    descriptions of simplicity and straightforwardness

    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
    件
    -0.14
    uars
    -0.14
     Seamless
    -0.14
    ูม
    -0.14
     Hos
    -0.13
    ivated
    -0.13
    pez
    -0.13
    rias
    -0.13
    ITIES
    -0.13
    obao
    -0.13
    POSITIVE LOGITS
     yet
    0.34
    yet
    0.29
    Yet
    0.25
     straightforward
    0.24
     Yet
    0.24
     elegance
    0.24
     elegant
    0.24
    ton
    0.21
    /basic
    0.21
     direct
    0.20
    Activations Density 0.020%

    No Known Activations