© 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. 15696
    Prev
    Next
    INDEX
    Explanations

    instances of time or duration references

    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
    abay
    -0.18
    stu
    -0.14
    ogan
    -0.14
    avy
    -0.13
    907
    -0.13
    ÄįÃŃ
    -0.13
    anten
    -0.13
     anale
    -0.13
    ruc
    -0.13
    ìļ´ëį°
    -0.13
    POSITIVE LOGITS
     last
    0.59
     finally
    0.52
    last
    0.48
    finally
    0.43
    -last
    0.42
     Finally
    0.39
     Lastly
    0.38
    æľĢåIJİ
    0.38
     LAST
    0.38
    lieÃŁlich
    0.37
    Activations Density 0.010%

    No Known Activations