© 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. 29699
    Prev
    Next
    INDEX
    Explanations

    instances of the word "nice" and its variations

    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
    sWith
    -0.19
    /Branch
    -0.19
    наÑĩе
    -0.17
    reach
    -0.17
    nds
    -0.16
    no
    -0.16
    swith
    -0.15
    OM
    -0.15
    tes
    -0.15
    utz
    -0.15
    POSITIVE LOGITS
    olson
    0.21
    olas
    0.17
    -looking
    0.17
    zsche
    0.17
    ptune
    0.16
    ER
    0.16
    lessly
    0.16
     Nic
    0.16
     sentiments
    0.16
    lsen
    0.15
    Activations Density 0.018%

    No Known Activations