© 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. 40607
    Prev
    Next
    INDEX
    Explanations

    expressions of personal reflection and judgment

    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
     indem
    -0.15
    .DropDown
    -0.15
    ebek
    -0.15
    PELL
    -0.15
    modifiable
    -0.14
     tÃŃm
    -0.14
    elay
    -0.14
    rug
    -0.14
    ãĥ¬ãĥ¼
    -0.14
     ëĶ°
    -0.14
    POSITIVE LOGITS
     given
    0.94
    given
    0.79
     considering
    0.77
     Given
    0.73
    Given
    0.68
     GIVEN
    0.61
     Considering
    0.59
    Considering
    0.56
    _given
    0.52
     since
    0.48
    Activations Density 0.077%

    No Known Activations