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. 29947
    Prev
    Next
    INDEX
    Explanations

    references to time-related phrases or actions

    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
    ercul
    -0.19
     Duy
    -0.14
    çĿ£
    -0.14
    _Execute
    -0.14
    oufl
    -0.13
     (!((
    -0.13
    ibles
    -0.13
    ัà¸ģร
    -0.13
    DMIN
    -0.13
    earable
    -0.13
    POSITIVE LOGITS
     done
    1.14
     Done
    1.02
    done
    0.97
     DONE
    0.93
    Done
    0.92
     finished
    0.83
    .done
    0.82
    -done
    0.81
    _done
    0.78
    DONE
    0.76
    Activations Density 0.035%

    No Known Activations