© 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. 9177
    Prev
    Next
    INDEX
    Explanations

    phrases related to political models and comparisons

    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
    .cod
    -0.14
    icari
    -0.13
    .enumer
    -0.12
    userinfo
    -0.12
    bracht
    -0.12
    cht
    -0.12
    raphics
    -0.12
    γη
    -0.12
    ืà¹ī
    -0.12
    Markup
    -0.12
    POSITIVE LOGITS
     example
    0.63
     model
    0.61
     examples
    0.56
    example
    0.53
    model
    0.52
     models
    0.50
     Model
    0.49
     Example
    0.49
     exemple
    0.48
    -example
    0.48
    Activations Density 0.036%

    No Known Activations