© 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 49152
    9. 8-RES_FS49152-JB
    11. 45552
    Prev
    Next
    INDEX
    Explanations

    specific numerical or quantified references

    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
    458
    +0.13
    0.6%
    700
    +0.12
    0.5%
    130
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    409
    +0.13
    0.02
    130
    +0.12
    0.02
    93
    +0.11
    0.02
    Negative Logits
    λ
    -0.90
    aline
    -0.88
    366
    -0.86
    MED
    -0.82
    vim
    -0.82
    姫
    -0.81
    vu
    -0.81
    ZZ
    -0.80
    essim
    -0.79
    aley
    -0.77
    POSITIVE LOGITS
     HOUSE
    1.10
     House
    1.08
     Houses
    0.98
     house
    0.93
    House
    0.90
    house
    0.90
     Sabha
    0.90
     Hann
    0.86
     King
    0.83
     houses
    0.83
    Activations Density 0.115%

    No Known Activations