© 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. 33041
    Prev
    Next
    INDEX
    Explanations

    references to performance metrics and their evaluation in various contexts

    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
    urple
    -0.17
     shorter
    -0.17
    opal
    -0.15
    usto
    -0.15
    è·Ŀ
    -0.14
     slower
    -0.14
     Continued
    -0.14
    à¹Ģà¸ľ
    -0.14
     Lev
    -0.14
     sat
    -0.13
    POSITIVE LOGITS
     increase
    0.43
     increases
    0.41
     increased
    0.41
     aument
    0.37
     Increase
    0.37
    incre
    0.35
    increase
    0.34
     Increases
    0.34
     Increased
    0.33
     increasing
    0.33
    Activations Density 0.051%

    No Known Activations