© 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. 54344
    Prev
    Next
    INDEX
    Explanations

    instances of the word "over" and its variations, indicating discussions on frequency or repetition

    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
    ikit
    -0.16
    vern
    -0.15
    atat
    -0.15
    리카
    -0.14
    év
    -0.14
     overview
    -0.14
    à¸ķา
    -0.14
    lis
    -0.14
    orian
    -0.13
    Overrides
    -0.13
    POSITIVE LOGITS
    tures
    0.25
    ture
    0.23
    lying
    0.20
    views
    0.20
    lord
    0.20
    leaf
    0.19
    stock
    0.19
    tones
    0.19
     course
    0.19
    ha
    0.18
    Activations Density 0.073%

    No Known Activations