© 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. 42170
    Prev
    Next
    INDEX
    Explanations

    interrogative phrases that seek explanations or clarifications

    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
    ulumi
    -0.16
    yen
    -0.16
    รà¸Ķ
    -0.14
    anche
    -0.14
    opia
    -0.14
    aeda
    -0.14
    asan
    -0.14
    .pixel
    -0.14
    ë¨
    -0.14
    ัà¸ĩà¹Ħม
    -0.14
    POSITIVE LOGITS
     much
    0.31
     exactly
    0.30
     long
    0.25
    much
    0.24
     far
    0.24
     best
    0.22
    -long
    0.22
    Much
    0.21
     soon
    0.21
    ls
    0.21
    Activations Density 0.060%

    No Known Activations