© 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. 63353
    Prev
    Next
    INDEX
    Explanations

    negative qualifiers and contrasting phrases

    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
    çĢ
    -0.13
    æk
    -0.13
    ̣
    -0.13
    éĴ®
    -0.13
    ï¸ı
    -0.12
    .hu
    -0.12
    ÅĽci
    -0.12
    ÐĽÐĺ
    -0.12
    گاÙĩ
    -0.12
     geil
    -0.12
    POSITIVE LOGITS
     how
    0.14
     that
    0.14
     any
    0.14
     would
    0.14
     what
    0.14
    illis
    0.13
     is
    0.13
     nor
    0.13
     daytime
    0.13
     does
    0.12
    Activations Density 0.003%

    No Known Activations