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

    Neuronpedia

    Jacobian LensNEW
    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 49152
    9. 8-RES_FS49152-JB
    11. 43000
    Prev
    Next
    INDEX
    Explanations

    different forms of a specific linguistic character

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    453
    +0.11
    0.5%
    35
    +0.11
    0.5%
    463
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.11
    0.01
    675
    +0.11
    0.01
    28
    +0.10
    0.01
    Negative Logits
    eday
    -0.84
    vironment
    -0.77
     Beir
    -0.77
    ĸļ
    -0.75
    hran
    -0.74
    achus
    -0.74
    bourg
    -0.73
     Granger
    -0.72
    ijah
    -0.71
    umerable
    -0.71
    POSITIVE LOGITS
    âķIJâķIJ
    1.21
    âķIJ
    1.15
    ा
    1.00
    quote
    0.95
    OUT
    0.95
    à¥
    0.93
    ÃĽ
    0.93
    Ķ
    0.88
    cffffcc
    0.86
    Ĺ
    0.85
    Activations Density 0.006%

    No Known Activations