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. 21813
    Prev
    Next
    INDEX
    Explanations

    numerical mentions of the number 16

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the number 16

    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
    •Embedding Space Digit Features

    No Comments

    Negative Logits
    taboola
    -0.75
    webkit
    -0.63
     Leban
    -0.63
    isner
    -0.62
    ffic
    -0.61
     appe
    -0.60
    rons
    -0.60
     aggro
    -0.59
     è£ıè
    -0.59
    heit
    -0.59
    POSITIVE LOGITS
     16
    2.84
     17
    2.18
     18
    1.95
     14
    1.94
     32
    1.93
    16
    1.86
     sixteen
    1.85
     19
    1.80
     15
    1.78
     13
    1.74
    Activations Density 0.049%

    No Known Activations