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. 3-RES-JB
    6. 7649
    Prev
    Next
    INDEX
    Explanations

    specific mentions of the word "statistic" in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to statisticians and statistical terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lungs
    -0.74
     palate
    -0.71
     curly
    -0.63
    perty
    -0.63
    hammad
    -0.63
     Sands
    -0.61
    uggage
    -0.61
     poisonous
    -0.61
     schooling
    -0.61
     acidic
    -0.60
    POSITIVE LOGITS
    ians
    1.32
    ian
    1.20
    inals
    1.10
    ially
    0.95
    iano
    0.93
    iel
    0.92
    istically
    0.90
    inal
    0.90
    ries
    0.88
    iani
    0.88
    Activations Density 0.011%

    No Known Activations