© 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. 54795
    Prev
    Next
    INDEX
    Explanations

    instances of the word "So" used to initiate statements or transitions in thought

    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
    dition
    -0.15
     so
    -0.14
    panies
    -0.14
    uty
    -0.13
    ttp
    -0.13
    nock
    -0.13
    _FATAL
    -0.13
    _SDK
    -0.13
    AZY
    -0.13
    ugar
    -0.13
    POSITIVE LOGITS
    oner
    0.30
    -called
    0.25
     far
    0.23
    aked
    0.23
    aring
    0.22
    apy
    0.21
    aking
    0.21
     yeah
    0.20
     here
    0.20
     yes
    0.20
    Activations Density 0.067%

    No Known Activations