© 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. 8-RES-JB
    11. 3328
    Prev
    Next
    INDEX
    Explanations

    phrases related to future timeframes

    oai_token-act-pair · gpt-3.5-turbo

    mentions of time frames related to 'next' periods, particularly in relation to years and decades

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Franch
    -0.79
    lees
    -0.71
    elli
    -0.69
    auer
    -0.68
    ioned
    -0.65
    ini
    -0.64
    had
    -0.63
    wright
    -0.62
    then
    -0.62
    acers
    -0.61
    POSITIVE LOGITS
     decade
    1.13
     millenn
    1.06
     fortnight
    0.95
     month
    0.90
     generation
    0.90
     installment
    0.89
    ebin
    0.89
     few
    0.86
     couple
    0.82
     week
    0.82
    Activations Density 0.044%

    No Known Activations