© 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. 45649
    Prev
    Next
    INDEX
    Explanations

    references to testimonies or statements made by individuals

    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
    713
    +0.11
    0.5%
    322
    +0.11
    0.5%
    103
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.11
    0.04
    566
    +0.11
    0.05
    354
    +0.10
    0.04
    Negative Logits
     unpop
    -0.72
     havoc
    -0.69
     overpowered
    -0.64
     nudity
    -0.63
     combustion
    -0.63
     unregulated
    -0.62
    âķIJ
    -0.60
     hotter
    -0.60
    yz
    -0.60
     recess
    -0.59
    POSITIVE LOGITS
     received
    0.91
     levied
    0.86
     issued
    0.86
     glean
    0.83
     uttered
    0.83
     relayed
    0.82
     devised
    0.82
     presented
    0.79
     undertook
    0.77
     forwarded
    0.77
    Activations Density 0.131%

    No Known Activations