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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    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. 34
    Prev
    Next
    INDEX
    Explanations

    references to historical monuments and infrastructure projects

    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
    оÑĢÑĤÑĥ
    -0.14
    zone
    -0.14
    eson
    -0.14
    990
    -0.14
    streams
    -0.14
    SPA
    -0.14
    út
    -0.13
    ony
    -0.13
    ÑĥÑģÑĤа
    -0.13
    zzo
    -0.13
    POSITIVE LOGITS
     py
    0.29
     Taj
    0.27
     Statue
    0.27
     Py
    0.26
     Notre
    0.26
     Tower
    0.24
     Trev
    0.24
     Lou
    0.23
    Py
    0.23
     Shard
    0.23
    Activations Density 0.076%

    No Known Activations