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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 11
    Prev
    Next
    INDEX
    Explanations

    themes of personal strength and resilience

    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
    IFO
    -0.14
    uries
    -0.14
    ]âĢı
    -0.14
    ÑĢоÑĩ
    -0.14
    ECTOR
    -0.14
    Occurs
    -0.14
    ILLA
    -0.13
    uitar
    -0.13
    akat
    -0.13
    retched
    -0.13
    POSITIVE LOGITS
     makes
    0.25
     definitely
    0.24
     is
    0.22
     certainly
    0.22
     speaks
    0.22
     means
    0.21
     make
    0.19
    à¸Ĺำà¹ĥห
    0.19
    åŃIJãģ¯
    0.18
     gives
    0.17
    Activations Density 0.312%

    No Known Activations