© 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. 22457
    Prev
    Next
    INDEX
    Explanations

    phrases related to superlatives and significant events

    oai_token-act-pair · gpt-3.5-turbo

    references to significant events or achievements

    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
    ĪĴ
    -0.73
    aleb
    -0.71
    eneg
    -0.68
    actionDate
    -0.67
    iencies
    -0.66
    milo
    -0.62
    letters
    -0.62
    gradient
    -0.61
    URL
    -0.61
    ¬¼
    -0.60
    POSITIVE LOGITS
     EVER
    1.07
     ever
    1.02
    ever
    0.97
     since
    0.83
     imaginable
    0.82
     anywhere
    0.74
    !.
    0.73
     shortest
    0.71
     besides
    0.69
    *.
    0.69
    Activations Density 0.481%

    No Known Activations