© 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. 2-RES-JB
    11. 1002
    Prev
    Next
    INDEX
    Explanations

    the word "lid"

    oai_token-act-pair · gpt-4-turbo

    words related to containers and their covers

    oai_token-act-pair · gpt-3.5-turbo

    references to lids and their various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WER
    -0.70
     Capital
    -0.69
    merce
    -0.69
    fortune
    -0.68
    bidden
    -0.68
     Bread
    -0.67
    Express
    -0.65
     Borough
    -0.65
    present
    -0.65
    Ĥª
    -0.65
    POSITIVE LOGITS
     lid
    1.37
    ysis
    0.97
    bell
    0.85
    wcs
    0.84
     curtain
    0.83
    achev
    0.82
    wheel
    0.82
    stones
    0.80
    bard
    0.80
     curtains
    0.77
    Activations Density 0.005%

    No Known Activations