© 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. 30164
    Prev
    Next
    INDEX
    Explanations

    instances of positive attributes or sentiments expressed in the text

    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
    ť
    -0.18
    osterone
    -0.18
    atik
    -0.16
    andler
    -0.16
    elts
    -0.15
    .GetInstance
    -0.15
    ucas
    -0.15
    áÅĻe
    -0.15
    ataire
    -0.15
    attern
    -0.14
    POSITIVE LOGITS
     news
    0.34
    news
    0.27
    News
    0.24
    -news
    0.24
     News
    0.24
     old
    0.23
     NEWS
    0.23
    .news
    0.21
     folks
    0.21
     thing
    0.19
    Activations Density 0.012%

    No Known Activations