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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 23368
    Prev
    Next
    INDEX
    Explanations

    mentions of financial resources or assets

    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
    eeee
    -0.19
    eous
    -0.18
    eid
    -0.18
    zzo
    -0.18
    eee
    -0.18
    oshi
    -0.17
    eam
    -0.17
    oÄŁ
    -0.17
    eed
    -0.16
    enuous
    -0.16
    POSITIVE LOGITS
    (es
    0.41
    s
    0.35
    mere
    0.25
    ers
    0.24
    iness
    0.24
    sar
    0.23
    ier
    0.22
    hh
    0.22
    es
    0.21
    sheets
    0.21
    Activations Density 0.024%

    No Known Activations