© 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. 8-RES-JB
    11. 755
    Prev
    Next
    INDEX
    Explanations

    mentions of winners or winning in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to winners in various contexts, particularly in competitions or contests

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uli
    -0.72
    ohyd
    -0.67
    senal
    -0.66
    heny
    -0.65
     probing
    -0.63
    unda
    -0.61
     sulf
    -0.60
    resent
    -0.60
     clamp
    -0.59
    Ku
    -0.58
    POSITIVE LOGITS
     winner
    0.89
     Winner
    0.84
     contestant
    0.83
     winners
    0.83
     prizes
    0.82
     prize
    0.80
    Winner
    0.79
    paces
    0.76
     contender
    0.73
     bidder
    0.73
    Activations Density 0.041%

    No Known Activations