© 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. 8-RES-JB
    11. 23732
    Prev
    Next
    INDEX
    Explanations

    references to sports achievements, especially those related to leading a team to finals or championships

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating leadership or guiding actions in a sports context

    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
    lus
    -0.66
    é¾įå¥ij士
    -0.64
     advised
    -0.62
    phalt
    -0.62
     compan
    -0.61
    chrom
    -0.60
     Affect
    -0.60
     speculated
    -0.58
    ulum
    -0.57
    accompanied
    -0.57
    POSITIVE LOGITS
     victory
    1.20
     victories
    1.11
     relegation
    0.92
     wins
    0.90
     undefeated
    0.89
     unbeat
    0.87
    iership
    0.87
     win
    0.83
     unbeaten
    0.83
     rebound
    0.79
    Activations Density 0.156%

    No Known Activations