© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 40819
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "no longer" in relation to personal or social status changes

    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
     yet
    -0.17
    yet
    -0.16
    enties
    -0.16
    Specifier
    -0.15
     already
    -0.15
     Yet
    -0.15
    ifton
    -0.15
    already
    -0.14
    899
    -0.14
    _lot
    -0.14
    POSITIVE LOGITS
     anymore
    0.38
     merely
    0.24
     exclusively
    0.21
     solely
    0.20
     nữa
    0.19
     simply
    0.19
     necessarily
    0.19
    éĤ£ä¹Ī
    0.18
     able
    0.18
    -exclusive
    0.18
    Activations Density 0.050%

    No Known Activations