© 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. 29251
    Prev
    Next
    INDEX
    Explanations

    instances of the word "is" and other forms of 'to be'

    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
    éĻ¢
    -0.15
     Appropri
    -0.14
    -controlled
    -0.14
    TexParameter
    -0.14
     controlled
    -0.13
    ENDIF
    -0.13
    TEGER
    -0.13
    mainwindow
    -0.13
    _procs
    -0.12
    uti
    -0.12
    POSITIVE LOGITS
     meant
    0.70
     intended
    0.50
     means
    0.33
     designed
    0.33
     mean
    0.31
     aimed
    0.30
    means
    0.28
     Means
    0.28
     intend
    0.27
    Means
    0.26
    Activations Density 0.152%

    No Known Activations