© 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. 29511
    Prev
    Next
    INDEX
    Explanations

    conjunctions and transitional phrases indicating comparisons or contrasts

    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
    aped
    -0.16
    pto
    -0.15
    ÙĪØ·
    -0.15
    roma
    -0.15
    OCUS
    -0.15
     |_|
    -0.15
    ubat
    -0.14
    ucc
    -0.14
    arium
    -0.14
     ÑĢаÑħ
    -0.14
    POSITIVE LOGITS
     atleast
    0.78
    èĩ³å°ij
    0.73
     least
    0.72
     alespoÅĪ
    0.64
    least
    0.61
     Least
    0.59
     minimum
    0.54
    Least
    0.51
     ØŃداÙĤÙĦ
    0.51
     ÑħоÑĤÑı
    0.48
    Activations Density 0.033%

    No Known Activations