© 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. 6493
    Prev
    Next
    INDEX
    Explanations

    words related to initiating or triggering events

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "sparked" and its variations in the context of events or actions that incite a significant reaction or change

    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
     avail
    -0.70
     Bastard
    -0.66
     Figure
    -0.63
     Appendix
    -0.62
     Tro
    -0.61
     Accounting
    -0.60
    onel
    -0.59
     Duty
    -0.57
     know
    -0.57
     li
    -0.57
    POSITIVE LOGITS
     sparked
    3.29
     ignited
    2.46
     provoked
    2.32
     spurred
    2.28
     triggered
    2.16
     stoked
    2.09
     prompted
    2.09
     fuelled
    2.02
     stirred
    1.94
     fueled
    1.86
    Activations Density 0.027%

    No Known Activations