© 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. 42337
    Prev
    Next
    INDEX
    Explanations

    comparisons of crime statistics between the US and other developed nations

    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
    urus
    -0.15
    unitOfWork
    -0.14
    StateException
    -0.14
     ActionTypes
    -0.14
    /bower
    -0.14
    ÙĦاÙĦ
    -0.13
    à¹ģà¸Ĥ
    -0.13
    Dash
    -0.13
    ÏĦÏī
    -0.13
    oningen
    -0.13
    POSITIVE LOGITS
     developed
    0.75
     Developed
    0.63
     advanced
    0.49
    develop
    0.47
    -develop
    0.45
    advanced
    0.42
    Develop
    0.39
     Advanced
    0.37
     industrial
    0.36
     Western
    0.36
    Activations Density 0.098%

    No Known Activations