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. 0-RES-JB
    6. 16285
    Prev
    Next
    INDEX
    Explanations

    words ending in the suffix "ily."

    oai_token-act-pair · gpt-3.5-turbo

    adverbs that end in 'ly'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Station
    -0.72
     Championship
    -0.64
     Grill
    -0.60
     PAT
    -0.56
     Brand
    -0.56
     Tuc
    -0.56
     ET
    -0.55
     metric
    -0.54
     protocol
    -0.54
     Index
    -0.54
    POSITIVE LOGITS
    ily
    4.64
    ILY
    2.47
    iness
    2.02
    iest
    1.95
    ilies
    1.74
    ier
    1.73
    lessly
    1.62
    ously
    1.61
    ishly
    1.53
    uously
    1.42
    Activations Density 0.009%

    No Known Activations