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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41771
    Prev
    Next
    INDEX
    Explanations

    contexts or phrases that indicate a sequence of actions or steps being taken

    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
     now
    -0.35
    now
    -0.25
     ahora
    -0.23
     ÑĤепеÑĢÑĮ
    -0.22
     NOW
    -0.21
    .now
    -0.21
     maintenant
    -0.20
     agora
    -0.20
    [now
    -0.20
    çİ°åľ¨
    -0.19
    POSITIVE LOGITS
    adays
    0.24
     onto
    0.24
     let
    0.23
     Ont
    0.21
     lets
    0.20
    let
    0.20
    onder
    0.20
    aday
    0.19
    Ont
    0.19
    onto
    0.18
    Activations Density 0.030%

    No Known Activations