© 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. 8169
    Prev
    Next
    INDEX
    Explanations

    verbs related to providing or explaining information, especially in a structured or procedural manner

    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
    332
    +0.13
    0.6%
    695
    +0.11
    0.5%
    60
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    332
    +0.13
    0.02
    60
    +0.11
    0.01
    695
    +0.11
    0.01
    Negative Logits
     Kubrick
    -0.70
     highlighting
    -0.65
    olson
    -0.63
    ãĥ£
    -0.63
     prone
    -0.62
     Revised
    -0.62
     Pound
    -0.60
     skating
    -0.59
     heavier
    -0.58
     Malta
    -0.57
    POSITIVE LOGITS
    iced
    1.22
    icing
    1.01
    itude
    1.01
    ility
    0.99
    serv
    0.97
    iments
    0.94
    ando
    0.93
    itors
    0.92
    maid
    0.91
    itary
    0.90
    Activations Density 0.017%

    No Known Activations