© 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. 63363
    Prev
    Next
    INDEX
    Explanations

    words indicating errors, lack, or an overwhelming number of items or issues

    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
    izations
    -0.15
    OUCH
    -0.15
    eson
    -0.15
    istique
    -0.14
    erin
    -0.14
    rak
    -0.14
     antics
    -0.14
    é©
    -0.14
    à¸Ńย
    -0.14
    trak
    -0.14
    POSITIVE LOGITS
    (sp
    0.25
    äºİ
    0.17
     vitae
    0.17
    æĸ¼
    0.16
    vatel
    0.16
    urate
    0.16
    ous
    0.15
    reamble
    0.14
    eniable
    0.14
    emente
    0.14
    Activations Density 0.173%

    No Known Activations