© 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. 23983
    Prev
    Next
    INDEX
    Explanations

    references to money or financial elements

    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
    ike
    -0.16
    abelle
    -0.15
    ousel
    -0.15
    .weixin
    -0.15
    .Encoding
    -0.15
    scape
    -0.14
    ackson
    -0.14
     fak
    -0.14
    iner
    -0.14
     Dove
    -0.14
    POSITIVE LOGITS
    ingham
    0.24
    minster
    0.23
    eye
    0.22
    aroo
    0.21
    etch
    0.20
    ered
    0.19
    ets
    0.19
    apest
    0.18
    eniable
    0.17
    led
    0.17
    Activations Density 0.014%

    No Known Activations