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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 23777
    Prev
    Next
    INDEX
    Explanations

    economic indicators related to debt and financial growth

    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
    -Za
    -0.16
     Setter
    -0.15
     Ferd
    -0.15
    Äģn
    -0.15
    óng
    -0.14
    day
    -0.14
    aln
    -0.13
     Five
    -0.13
     Snape
    -0.13
    tte
    -0.13
    POSITIVE LOGITS
     Q
    0.53
     fourth
    0.45
     third
    0.44
    Q
    0.44
     second
    0.42
     q
    0.36
    第äºĮ
    0.35
    ç¬¬åĽĽ
    0.35
    third
    0.35
     quarter
    0.34
    Activations Density 0.039%

    No Known Activations