© 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. 0-RES-JB
    11. 4032
    Prev
    Next
    INDEX
    Explanations

    countable nouns preceded by words indicating a large quantity, such as "Thousands" and "Hundreds"

    oai_token-act-pair · gpt-3.5-turbo

    references to large groups of people, specifically thousands or hundreds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     implied
    -0.77
     otherwise
    -0.74
     salad
    -0.72
     propri
    -0.71
     eval
    -0.69
     evaluation
    -0.68
     consistency
    -0.68
     regression
    -0.67
     homework
    -0.67
     overall
    -0.67
    POSITIVE LOGITS
    Thousands
    3.00
    Hundreds
    2.68
     Thousands
    2.42
     Hundreds
    2.23
     Dozens
    1.82
     Millions
    1.68
     Tens
    1.45
    ousands
    1.34
    zens
    1.34
    undreds
    1.29
    Activations Density 0.029%

    No Known Activations