Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 19910
    Prev
    Next
    INDEX
    Explanations

    words related to accuracy or being correct

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "accurate."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NetMessage
    -0.88
    hov
    -0.84
    Rus
    -0.79
    neys
    -0.79
    ATA
    -0.74
    cheon
    -0.74
    hibit
    -0.73
    LAN
    -0.73
    spl
    -0.73
    ETF
    -0.73
    POSITIVE LOGITS
     accurate
    1.27
    uracy
    1.17
     inaccurate
    1.16
     accuracy
    1.16
     inacc
    1.14
     portrayal
    1.02
     depiction
    1.00
     inaccur
    0.97
     accurately
    0.94
     truthful
    0.92
    Activations Density 0.006%

    No Known Activations