© 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. 9122
    Prev
    Next
    INDEX
    Explanations

    words related to the process of becoming or transformation

    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
    ished
    -0.18
    ted
    -0.17
    geries
    -0.16
    eum
    -0.16
    eced
    -0.16
    visor
    -0.15
    ulence
    -0.15
    ÑĨик
    -0.15
    ishment
    -0.15
    iable
    -0.14
    POSITIVE LOGITS
    oming
    0.29
    omes
    0.27
    ing
    0.22
    ame
    0.21
    ome
    0.20
    ause
    0.19
    autiful
    0.19
    come
    0.19
    COME
    0.17
    aucoup
    0.17
    Activations Density 0.005%

    No Known Activations