© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 11981
    Prev
    Next
    INDEX
    Explanations

    sports-related terms and player names

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    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

    Neuron Alignment
    Index
    Value
    % of L₁
    178
    +0.12
    0.5%
    2
    +0.10
    0.5%
    420
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    406
    +0.12
    0.02
    404
    +0.10
    0.02
    464
    +0.10
    0.02
    Negative Logits
    byss
    -0.71
    Ô
    -0.70
     Witcher
    -0.69
    shown
    -0.65
    ikuman
    -0.63
     Patient
    -0.63
    ð
    -0.62
     Tsukuyomi
    -0.62
     reconc
    -0.61
    è£ħ
    -0.61
    POSITIVE LOGITS
    TING
    1.05
    tered
    1.01
    tering
    1.00
    ches
    1.00
    ted
    0.99
    ters
    0.98
    tle
    0.95
    wana
    0.93
    ting
    0.92
    hod
    0.89
    Activations Density 0.032%

    No Known Activations