Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 2627
    Prev
    Next
    INDEX
    Explanations

    phrases related to politics and power dynamics

    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
    isan
    -0.19
    éĨ
    -0.15
    ikat
    -0.15
    lernen
    -0.15
    figcaption
    -0.14
    oci
    -0.14
    ullen
    -0.14
    æŁ³
    -0.14
    oping
    -0.14
    earch
    -0.14
    POSITIVE LOGITS
     cl
    0.43
     standing
    0.42
     muscle
    0.36
     Standing
    0.34
    Standing
    0.31
    standing
    0.30
     Muscle
    0.30
     bona
    0.28
     staying
    0.27
     chops
    0.26
    Activations Density 0.046%

    No Known Activations