© 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. 29685
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "more" in various contexts

    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
    æĽ´å¤ļ
    -0.18
    ysa
    -0.17
    ÑģÑı
    -0.16
     Otherwise
    -0.15
     anymore
    -0.14
    àµį
    -0.14
     lots
    -0.14
     OTHERWISE
    -0.14
    ikel
    -0.14
    ़
    -0.14
    POSITIVE LOGITS
     than
    0.61
    -than
    0.49
    than
    0.46
     THAN
    0.40
    _than
    0.37
     Than
    0.35
     niż
    0.33
    Than
    0.33
     än
    0.32
     než
    0.29
    Activations Density 0.059%

    No Known Activations