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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1227
    Prev
    Next
    INDEX
    Explanations

    descriptions of physical interactions and movements

    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
    87
    +0.31
    1.7%
    678
    +0.12
    0.7%
    570
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.31
    -0.03
    570
    +0.12
    0.03
    678
    +0.10
    0.07
    Negative Logits
    ylum
    -0.77
     Founders
    -0.75
     Founding
    -0.74
     Comprehensive
    -0.71
    itutional
    -0.71
     internationally
    -0.70
     charities
    -0.67
     Markets
    -0.67
     Malays
    -0.67
     Acad
    -0.66
    POSITIVE LOGITS
     crou
    0.99
     blinking
    0.95
     eject
    0.93
     vibration
    0.92
     velocity
    0.90
     bounces
    0.89
     vibrations
    0.89
     flick
    0.88
     audible
    0.87
     braking
    0.86
    Activations Density 1.968%

    No Known Activations