© 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 768
    9. 8-RES_FS768-JB
    11. 752
    Prev
    Next
    INDEX
    Explanations

    dates and significant events from sports-related texts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    146
    +0.12
    0.6%
    306
    +0.11
    0.5%
    598
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    306
    +0.12
    0.16
    146
    +0.11
    0.15
    689
    +0.10
    0.14
    Negative Logits
     ancest
    -0.74
    ube
    -0.66
    hed
    -0.65
    este
    -0.62
    phy
    -0.61
    ype
    -0.59
    icate
    -0.58
     kin
    -0.58
    bars
    -0.57
    imore
    -0.57
    POSITIVE LOGITS
    å¹
    1.22
    -'
    1.14
     onwards
    1.10
     onward
    0.95
    âĢ²
    0.84
    20439
    0.80
     edition
    0.77
    âĢ³
    0.75
     census
    0.73
    –
    0.69
    Activations Density 1.786%

    No Known Activations