© 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 12288
    9. 8-RES_FS12288-JB
    11. 1636
    Prev
    Next
    INDEX
    Explanations

    dialogue and speech patterns

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    640
    +0.15
    0.7%
    6
    +0.14
    0.7%
    430
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.15
    0.10
    430
    +0.14
    0.07
    618
    +0.13
    0.07
    Negative Logits
    itionally
    -0.75
    isphere
    -0.73
    omics
    -0.72
     viable
    -0.72
    ital
    -0.72
    NW
    -0.72
     exclusively
    -0.71
     transition
    -0.70
     campaigned
    -0.70
    utsche
    -0.69
    POSITIVE LOGITS
     Suddenly
    1.44
     Slowly
    1.35
     Then
    1.25
     Someone
    1.22
     Somehow
    1.21
     Something
    1.18
     Seeing
    1.15
     Everyone
    1.13
     Thinking
    1.13
     Checking
    1.11
    Activations Density 0.313%

    No Known Activations