© 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. 41096
    Prev
    Next
    INDEX
    Explanations

    expressions of emotional reactions or evaluations

    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
     yes
    -0.14
     Yes
    -0.14
    halt
    -0.14
    à¹īà¸Ī
    -0.14
    aria
    -0.14
    omatic
    -0.14
     ÑģпÑĢÑı
    -0.13
     sill
    -0.13
    anza
    -0.13
    SEL
    -0.13
    POSITIVE LOGITS
     shr
    0.74
     shrugged
    0.66
    shr
    0.62
     Shr
    0.58
     SHR
    0.41
     shake
    0.32
     indifference
    0.31
     Shake
    0.31
     shrine
    0.30
     shrink
    0.29
    Activations Density 0.016%

    No Known Activations