© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 22692
    Prev
    Next
    INDEX
    Explanations

    technical terms related to fishing techniques and equipment

    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
     ocean
    -0.26
     oceans
    -0.26
     Ocean
    -0.25
    Ocean
    -0.23
     seas
    -0.23
     Seas
    -0.22
     coastal
    -0.21
    æµ·
    -0.21
     seaside
    -0.20
     sea
    -0.19
    POSITIVE LOGITS
     fish
    0.30
     aquatic
    0.29
     floating
    0.28
     bank
    0.27
     water
    0.26
     Floating
    0.26
     float
    0.25
     ducks
    0.25
     algae
    0.24
     Fish
    0.24
    Activations Density 0.026%

    No Known Activations