© 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. 8-RES-JB
    11. 190
    Prev
    Next
    INDEX
    Explanations

    time-related expressions, specifically mentioning weeks, months, days, and years

    oai_token-act-pair · gpt-3.5-turbo

    references to time periods and durations

    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
    •single token firings
    strong activation on " month", " year", " week"

    No Comments

    Negative Logits
    reditary
    -0.72
    kef
    -0.70
     Borders
    -0.68
     predec
    -0.67
     arrang
    -0.64
     unnecess
    -0.62
     exha
    -0.61
     Mellon
    -0.60
    hur
    -0.58
     Strait
    -0.58
    POSITIVE LOGITS
     ago
    0.68
    cloud
    0.65
    view
    0.65
    dream
    0.64
    trip
    0.63
     ][
    0.61
    moon
    0.61
    coin
    0.60
     >=
    0.60
     elapsed
    0.58
    Activations Density 0.062%

    No Known Activations