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. 29835
    Prev
    Next
    INDEX
    Explanations

    phrases indicating public interest and support for initiatives or proposals

    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
    ameda
    -0.16
    alat
    -0.14
    ysa
    -0.14
    endencies
    -0.13
    roit
    -0.13
    jem
    -0.13
     Mediterr
    -0.13
    iec
    -0.13
     onBindViewHolder
    -0.13
     demokrat
    -0.13
    POSITIVE LOGITS
     support
    0.41
     momentum
    0.38
     Support
    0.32
    support
    0.30
     buy
    0.30
     SUPPORT
    0.29
     consensus
    0.28
    Support
    0.28
     interest
    0.28
     agreement
    0.26
    Activations Density 0.084%

    No Known Activations