© 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. 7072
    Prev
    Next
    INDEX
    Explanations

    words related to American football teams, such as team names and analysis terms

    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
    288
    +0.11
    0.5%
    282
    +0.10
    0.5%
    31
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    548
    +0.11
    0.01
    656
    +0.10
    0.02
    31
    +0.09
    0.02
    Negative Logits
    phas
    -0.94
    hig
    -0.83
    lasses
    -0.79
    pora
    -0.77
    nces
    -0.75
    perse
    -0.74
    selves
    -0.74
    arist
    -0.73
    uchin
    -0.72
    cam
    -0.72
    POSITIVE LOGITS
    sburgh
    1.65
     Steelers
    1.27
     Penguins
    1.22
    sylvania
    1.06
     Pirates
    0.94
    sburg
    0.91
    Pitt
    0.87
     Mellon
    0.84
    sylv
    0.81
     Pike
    0.80
    Activations Density 0.030%

    No Known Activations