© 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. 10133
    Prev
    Next
    INDEX
    Explanations

    numbers

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Number" followed by numerical values

    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
    pas
    -0.66
     eas
    -0.65
     goods
    -0.63
    kes
    -0.63
     passively
    -0.61
     plains
    -0.60
     stim
    -0.60
     bene
    -0.59
     rails
    -0.58
     convinc
    -0.57
    POSITIVE LOGITS
     Number
    3.58
    Number
    2.97
    number
    2.26
     Numbers
    2.03
     number
    1.89
     NUM
    1.85
    Numbers
    1.55
    NUM
    1.50
     numbers
    1.41
     Amount
    1.21
    Activations Density 0.014%

    No Known Activations