© 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. 11-RES-JB
    11. 15672
    Prev
    Next
    INDEX
    Explanations

    phrases related to pieces or components

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "piece" used to refer to components or elements of a larger whole

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Predators
    -0.66
    elsius
    -0.66
     Monitor
    -0.65
    iour
    -0.62
    acebook
    -0.60
    Fight
    -0.60
    rl
    -0.56
    few
    -0.56
    ESCO
    -0.55
     Nadu
    -0.54
    POSITIVE LOGITS
    meal
    2.03
    work
    1.10
    wise
    0.94
     thereof
    0.94
     of
    0.87
    umen
    0.87
    icles
    0.85
    toe
    0.80
     pieces
    0.79
    ridges
    0.79
    Activations Density 0.055%

    No Known Activations