© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 44947
    Prev
    Next
    INDEX
    Explanations

    economic indicators and terms related to housing and employment markets

    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
    à¹īà¸ĩ
    -0.16
    XX
    -0.14
    chner
    -0.14
     behalf
    -0.14
    à¸Ħว
    -0.13
    ÎķÎļ
    -0.13
    arella
    -0.13
    .px
    -0.12
    Ĥ
    -0.12
     ÑĢÑĥк
    -0.12
    POSITIVE LOGITS
     drop
    0.20
     figures
    0.20
     picture
    0.19
     current
    0.18
     recent
    0.18
     figure
    0.18
     latest
    0.18
     economy
    0.18
     outlook
    0.18
     combination
    0.18
    Activations Density 0.091%

    No Known Activations