© 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. 54792
    Prev
    Next
    INDEX
    Explanations

    high-frequency words that convey emphasis or quantity

    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
    aurus
    -0.18
    èĤĥ
    -0.14
    oras
    -0.13
    \controllers
    -0.13
     Trojan
    -0.13
    atus
    -0.13
     sophistication
    -0.13
    igg
    -0.13
    449
    -0.12
     Nobel
    -0.12
    POSITIVE LOGITS
     compared
    0.20
     overall
    0.18
    ëĮĢë¹Ħ
    0.17
     Kostenlos
    0.15
    ovice
    0.15
     Overall
    0.15
    edenÃŃ
    0.14
    overall
    0.14
    /msg
    0.14
    oins
    0.14
    Activations Density 0.000%

    No Known Activations