Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 31419
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions

    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
    .bmp
    -0.18
    ãģŁãĤģãģ®
    -0.15
    otope
    -0.14
    ich
    -0.14
    ORMAL
    -0.14
    .debian
    -0.14
    ibr
    -0.14
    apo
    -0.14
    compat
    -0.14
    ere
    -0.14
    POSITIVE LOGITS
     pace
    0.16
    argo
    0.16
    PACE
    0.15
     advantage
    0.15
     mercy
    0.15
    onec
    0.15
    пÑĢимеÑĢ
    0.15
     fault
    0.15
    INCLUDED
    0.14
     satisfaction
    0.14
    Activations Density 0.016%

    No Known Activations