© 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. 54387
    Prev
    Next
    INDEX
    Explanations

    references to star ratings or accolades related to dining experiences

    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
    ëłī
    -0.15
    ÏĦίοÏħ
    -0.14
     плен
    -0.13
    ansi
    -0.13
    Pragma
    -0.13
    ekyll
    -0.13
    ffi
    -0.13
    RuntimeObject
    -0.12
    νή
    -0.12
    é³´
    -0.12
    POSITIVE LOGITS
     star
    1.32
     stars
    1.23
    -star
    1.09
     Star
    1.07
    star
    1.07
     Stars
    1.04
    stars
    0.99
    Star
    0.97
    æĺŁ
    0.97
     STAR
    0.92
    Activations Density 0.010%

    No Known Activations