© 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. 14618
    Prev
    Next
    INDEX
    Explanations

    information related to film releases and their promotional materials

    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
     --
    -0.20
    ä»Ĭå¹´
    -0.19
     tonight
    -0.18
     yesterday
    -0.17
     today
    -0.17
    ообÑĢаз
    -0.17
     tomorrow
    -0.17
    åĪļæīį
    -0.15
     here
    -0.15
    ãģĭãģ£ãģ¦
    -0.15
    POSITIVE LOGITS
     stated
    0.18
    [c
    0.17
    ,[
    0.17
    {{
    0.17
    âĢ²
    0.17
    .[
    0.16
    201
    0.15
    âĢħ
    0.15
    ]]
    0.15
    [a
    0.15
    Activations Density 0.167%

    No Known Activations