© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 1701
    Prev
    Next
    INDEX
    Explanations

    numbers related to quantities or measurements

    oai_token-act-pair · gpt-3.5-turbo

    repeated numerical values, particularly "99" and "98"

    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
    endar
    -0.72
    culated
    -0.61
    NetMessage
    -0.61
    iber
    -0.60
    inson
    -0.59
    iciary
    -0.59
    ework
    -0.59
    interstitial
    -0.59
    BuyableInstoreAndOnline
    -0.58
     alleg
    -0.58
    POSITIVE LOGITS
     99
    3.28
     98
    2.58
     97
    2.30
     95
    2.15
     96
    1.91
     94
    1.87
     93
    1.82
     89
    1.71
     91
    1.68
     92
    1.67
    Activations Density 0.009%

    No Known Activations