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. 23526
    Prev
    Next
    INDEX
    Explanations

    compliments and expressions of satisfaction related to products or services

    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
    amient
    -0.18
     ogl
    -0.16
    ниÑĨÑĮ
    -0.15
     processData
    -0.15
    vro
    -0.15
    atel
    -0.14
    igram
    -0.14
    &view
    -0.14
    edith
    -0.14
    .Marker
    -0.14
    POSITIVE LOGITS
     done
    0.25
     managed
    0.23
     doing
    0.23
     manage
    0.22
    done
    0.22
     certainly
    0.22
     nail
    0.21
     succeeded
    0.20
    doing
    0.20
     sure
    0.20
    Activations Density 0.085%

    No Known Activations