© 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. 29503
    Prev
    Next
    INDEX
    Explanations

    references to significant or impactful events, particularly financial or political in nature

    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
    inski
    -0.17
    esian
    -0.15
     aggression
    -0.14
     Tome
    -0.14
    è¶Ĭ
    -0.14
    away
    -0.13
    orem
    -0.13
    bell
    -0.13
    insky
    -0.13
    aran
    -0.13
    POSITIVE LOGITS
     increase
    0.36
     decrease
    0.30
     rise
    0.29
    increase
    0.27
     Increase
    0.26
     decline
    0.24
    Increase
    0.24
     reduction
    0.24
     spike
    0.22
    _increase
    0.22
    Activations Density 0.058%

    No Known Activations