© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23835
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal beliefs and identity

    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
    ãĤ¡
    -0.15
    dre
    -0.15
     Scheme
    -0.15
    upt
    -0.14
    è¼Ŀ
    -0.14
    uilder
    -0.14
    .trace
    -0.14
    Scheme
    -0.14
     scheme
    -0.14
    oden
    -0.13
    POSITIVE LOGITS
     raised
    0.62
    raised
    0.53
     Raised
    0.52
    Raised
    0.44
     raise
    0.42
     taught
    0.41
    -ra
    0.38
     raises
    0.38
     raising
    0.37
     Raise
    0.36
    Activations Density 0.051%

    No Known Activations