© 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. 23590
    Prev
    Next
    INDEX
    Explanations

    phrases related to aftermath or consequences of events

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "wake," particularly in contexts related to events or incidents that follow significant occurrences

    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
     documentation
    -0.62
    venants
    -0.62
    remlin
    -0.61
     rib
    -0.61
     forbidden
    -0.60
     classified
    -0.60
    inates
    -0.60
    classified
    -0.59
     Rouge
    -0.59
    eq
    -0.59
    POSITIVE LOGITS
    wake
    1.37
     wake
    1.17
    mares
    1.13
     Wake
    1.03
    fulness
    1.02
    achine
    0.92
    Accessory
    0.83
    ãĥĹ
    0.80
    bringer
    0.79
    stead
    0.78
    Activations Density 0.006%

    No Known Activations