© 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. 2931
    Prev
    Next
    INDEX
    Explanations

    words related to competitive races or contests

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "race."

    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
     vou
    -0.69
     Vera
    -0.69
    ç«
    -0.68
    vor
    -0.64
    answered
    -0.64
    hern
    -0.64
    fired
    -0.62
     Prov
    -0.62
    Lay
    -0.62
    stairs
    -0.61
    POSITIVE LOGITS
     race
    3.93
     Race
    2.99
     races
    2.80
    Race
    2.78
    race
    2.64
     Races
    2.17
     racing
    1.70
     racial
    1.60
     racer
    1.57
     raced
    1.54
    Activations Density 0.017%

    No Known Activations