© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 40972
    Prev
    Next
    INDEX
    Explanations

    references to typical characteristics or qualities of films, particularly related to expectations versus reality

    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
    odal
    -0.14
    ynam
    -0.14
    _CD
    -0.14
    rum
    -0.14
    erald
    -0.13
     gameplay
    -0.13
     Cyril
    -0.13
    uma
    -0.13
    iferay
    -0.13
    izarre
    -0.13
    POSITIVE LOGITS
     movies
    0.34
     films
    0.33
     flick
    0.26
     feature
    0.24
     action
    0.23
     sequ
    0.22
     motion
    0.21
     prestige
    0.21
     Movies
    0.20
     Films
    0.20
    Activations Density 0.168%

    No Known Activations