© 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. 23978
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "in" used within 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
     tomorrow
    -0.21
    -years
    -0.21
     Tomorrow
    -0.20
    Tomorrow
    -0.18
     decade
    -0.18
     YEARS
    -0.17
    197
    -0.17
    200
    -0.17
    Years
    -0.17
    199
    -0.16
    POSITIVE LOGITS
     late
    0.67
    late
    0.51
     February
    0.48
     July
    0.47
     Late
    0.47
     early
    0.47
     January
    0.46
     November
    0.46
     June
    0.45
     October
    0.44
    Activations Density 0.063%

    No Known Activations