© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10-RES-JB
    11. 4518
    Prev
    Next
    INDEX
    Explanations

    dates in the format of month followed by a day, such as "August 3" or terms related to appointments or scheduling like "host" or "Add" with numbers following

    oai_token-act-pair · gpt-3.5-turbo

    specific dates and numerical references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aya
    -0.81
    ayan
    -0.81
    aches
    -0.77
     Augusta
    -0.76
     Flores
    -0.75
    anka
    -0.73
    uay
    -0.73
     Kemp
    -0.73
     Rowe
    -0.71
     Grace
    -0.69
    POSITIVE LOGITS
    3
    1.39
     3
    1.29
     III
    1.13
    three
    1.09
     03
    1.02
     Three
    1.02
    03
    1.00
    Third
    1.00
    Three
    0.97
     Third
    0.97
    Activations Density 0.334%

    No Known Activations