© 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. 41735
    Prev
    Next
    INDEX
    Explanations

    references to statistical relationships and findings in research studies

    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
     fate
    -0.16
    éľ²
    -0.15
    ymi
    -0.15
    ãģ¾ãĤĬ
    -0.14
     Fate
    -0.14
    eva
    -0.14
    fuse
    -0.14
    bol
    -0.13
    ãĤ¤ãĥī
    -0.13
     fait
    -0.13
    POSITIVE LOGITS
     association
    0.42
     associations
    0.39
    association
    0.34
     relationship
    0.33
    associ
    0.29
     Associations
    0.29
     relationships
    0.29
     link
    0.29
     Association
    0.28
     assoc
    0.27
    Activations Density 0.147%

    No Known Activations