© 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. 41294
    Prev
    Next
    INDEX
    Explanations

    phrases indicating ongoing projects or developments

    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
     finally
    -0.19
    ersist
    -0.18
     eventually
    -0.17
     ultimately
    -0.17
     NSAttributedString
    -0.15
    finally
    -0.15
    galement
    -0.15
     lagi
    -0.14
     ÙĤÙĬ
    -0.14
    íĥģ
    -0.14
    POSITIVE LOGITS
     Already
    0.41
     already
    0.40
    Already
    0.38
    already
    0.35
     begun
    0.34
     preliminary
    0.31
     beginning
    0.31
     begin
    0.31
    _already
    0.29
     Beginning
    0.28
    Activations Density 0.016%

    No Known Activations