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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6705
    Prev
    Next
    INDEX
    Explanations

    keywords related to sequences of events

    oai_token-act-pair · gpt-3.5-turbo

    sequences that introduce consequences or follow-up actions

    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
     outweigh
    -0.67
     sporting
    -0.63
     spelling
    -0.62
     caution
    -0.62
     dating
    -0.61
     merit
    -0.60
     offensive
    -0.59
     interests
    -0.59
     individuals
    -0.59
     individual
    -0.58
    POSITIVE LOGITS
     Then
    3.27
    Then
    2.78
     THEN
    2.44
    then
    2.38
     Afterwards
    2.20
     then
    1.95
     Later
    1.76
     Finally
    1.60
     Eventually
    1.56
     Otherwise
    1.48
    Activations Density 0.013%

    No Known Activations