© 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. 30196
    Prev
    Next
    INDEX
    Explanations

    technical terms related to programming and computation processes

    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
    æŀ¶
    -0.14
     whitelist
    -0.14
    /mit
    -0.14
     caffe
    -0.14
    CAST
    -0.14
    herits
    -0.14
     prox
    -0.13
    icap
    -0.13
     industrial
    -0.13
     Cas
    -0.13
    POSITIVE LOGITS
     resident
    0.20
    resident
    0.19
     routines
    0.18
     Resident
    0.16
    vä
    0.16
     routine
    0.16
     fetch
    0.16
     chá»ĵng
    0.16
    NZ
    0.15
     FETCH
    0.15
    Activations Density 0.023%

    No Known Activations