© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 41697
    Prev
    Next
    INDEX
    Explanations

    expressions of capability and potential

    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
    icens
    -0.18
    anca
    -0.17
    mÃŃt
    -0.16
    pton
    -0.16
    á»§ng
    -0.16
    меÑĤÑĮ
    -0.15
    .scalablytyped
    -0.15
    edom
    -0.14
    deÅŁ
    -0.14
    alfa
    -0.14
    POSITIVE LOGITS
     handle
    0.30
     handles
    0.28
     indeed
    0.27
    handle
    0.24
     HANDLE
    0.24
     Handle
    0.24
     Handles
    0.23
     handled
    0.22
    .handle
    0.22
     handling
    0.21
    Activations Density 0.096%

    No Known Activations