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

    Neuronpedia

    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. 2-RES-JB
    11. 5344
    Prev
    Next
    INDEX
    Explanations

    mentions of individuals or characters being described as champions or having championship-related attributes

    oai_token-act-pair · gpt-3.5-turbo

    references to champions in 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
    eret
    -0.81
    uld
    -0.77
    kefeller
    -0.73
    uggage
    -0.69
    aith
    -0.68
     ali
    -0.68
    gettable
    -0.67
    pora
    -0.67
    trak
    -0.66
    iday
    -0.66
    POSITIVE LOGITS
     champion
    1.07
     champions
    0.99
     Champion
    0.94
    ampion
    0.83
     boxer
    0.80
     challenger
    0.77
    iors
    0.74
     Chall
    0.74
    åĭ
    0.73
    naire
    0.71
    Activations Density 0.011%

    No Known Activations