© 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. 63283
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "the."

    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
    ึà¹Ī
    -0.14
    istik
    -0.14
    PX
    -0.14
    resa
    -0.14
    òng
    -0.14
    ukan
    -0.14
    iges
    -0.13
     pump
    -0.13
    ase
    -0.13
     cose
    -0.13
    POSITIVE LOGITS
     city
    0.67
     region
    0.59
     province
    0.53
    city
    0.49
    region
    0.45
     town
    0.40
     state
    0.40
     county
    0.39
    å¸Ĥ
    0.38
    -region
    0.37
    Activations Density 0.039%

    No Known Activations