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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 54579
    Prev
    Next
    INDEX
    Explanations

    terms related to compliance or submission

    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
    strar
    -0.17
    zy
    -0.16
    ulle
    -0.15
    911
    -0.15
    igid
    -0.15
     addCriterion
    -0.14
     ca
    -0.14
    achable
    -0.14
    156
    -0.14
    ule
    -0.14
    POSITIVE LOGITS
     acqu
    0.26
    ittal
    0.24
    ies
    0.21
    ần
    0.17
     acquaintance
    0.16
    isse
    0.16
    ãĥ¼ãĤº
    0.15
    quan
    0.15
    ¼
    0.15
    NSURLSession
    0.14
    Activations Density 0.004%

    No Known Activations