© 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. 32905
    Prev
    Next
    INDEX
    Explanations

    locations and military terminology

    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
    ichel
    -0.16
    ernals
    -0.15
     Miner
    -0.15
    á»ģn
    -0.15
    iquer
    -0.15
    _likelihood
    -0.15
    ãĤ¶
    -0.15
     spi
    -0.14
    arra
    -0.14
    izia
    -0.14
    POSITIVE LOGITS
     Camp
    0.33
     Joint
    0.32
     base
    0.31
     bases
    0.29
     Base
    0.29
     Naval
    0.29
    Joint
    0.26
    -base
    0.26
     Quant
    0.26
     Fort
    0.24
    Activations Density 0.040%

    No Known Activations