© 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. 29487
    Prev
    Next
    INDEX
    Explanations

    present and gerund verbs, particularly related to action or process

    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
    /remove
    -0.18
    chter
    -0.18
    quam
    -0.17
    chet
    -0.16
    chal
    -0.15
    ģm
    -0.15
    /close
    -0.15
    'gc
    -0.15
    ERTICAL
    -0.15
    ily
    -0.15
    POSITIVE LOGITS
    redients
    0.26
     oneself
    0.24
    redient
    0.23
     of
    0.22
     techniques
    0.22
    /logging
    0.22
     spree
    0.22
     technique
    0.21
    enuity
    0.21
    ton
    0.20
    Activations Density 0.296%

    No Known Activations