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

    Neuronpedia

    Jacobian LensNEW
    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. 42871
    Prev
    Next
    INDEX
    Explanations

    phrases related to packaging and delivery processes

    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
     Hairst
    -0.15
    byt
    -0.13
    anova
    -0.13
    j
    -0.13
    odal
    -0.13
     xsi
    -0.13
    ansson
    -0.13
    oeff
    -0.13
    ernal
    -0.13
    άÏĥ
    -0.13
    POSITIVE LOGITS
     wrap
    1.01
     wrapping
    0.96
     Wrap
    0.94
     wraps
    0.90
     wrapped
    0.89
    wrap
    0.89
    Wrap
    0.85
     Wr
    0.82
    -wrap
    0.81
     wrapper
    0.79
    Activations Density 0.017%

    No Known Activations