© 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. 1-RES-JB
    11. 11206
    Prev
    Next
    INDEX
    Explanations

    mentions of sporting events and achievements, especially related to singles and titles

    oai_token-act-pair · gpt-3.5-turbo

    references to singles and marital status

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Dra
    -0.71
    Downloadha
    -0.71
    anwhile
    -0.71
     Lup
    -0.66
     Rand
    -0.64
     Arbor
    -0.63
     Austral
    -0.61
     Randolph
    -0.61
     Establishment
    -0.61
     Greece
    -0.60
    POSITIVE LOGITS
     singles
    1.17
    peed
    0.98
    creen
    0.93
     crochet
    0.84
    bledon
    0.77
    peak
    0.76
    paces
    0.75
     challeng
    0.74
    sheet
    0.72
    ters
    0.72
    Activations Density 0.005%

    No Known Activations