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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 54392
    Prev
    Next
    INDEX
    Explanations

    references to statistical data and statistics in general

    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
    ing
    -0.17
    strict
    -0.17
    thon
    -0.16
    گاÙĩ
    -0.16
    æĬ¼
    -0.15
    tip
    -0.15
    sworth
    -0.15
    оÑĢалÑĮ
    -0.15
    erp
    -0.15
    tone
    -0.15
    POSITIVE LOGITS
    istical
    0.25
    itics
    0.22
     quo
    0.22
    /stat
    0.20
    ALLY
    0.18
    ISTICS
    0.18
    icians
    0.18
    vfs
    0.17
    istically
    0.17
    cipher
    0.17
    Activations Density 0.018%

    No Known Activations