© 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 1536
    9. 8-RES_FS1536-JB
    11. 75
    Prev
    Next
    INDEX
    Explanations

    information related to travels, locations, and arrangements of people and events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    37
    +0.11
    0.5%
    468
    +0.11
    0.5%
    208
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    468
    +0.11
    0.12
    69
    +0.11
    0.10
    37
    +0.11
    0.12
    Negative Logits
    selves
    -0.99
     unison
    -0.86
     selves
    -0.71
    hub
    -0.70
     respective
    -0.68
    results
    -0.64
     Authors
    -0.62
     respectively
    -0.62
    they
    -0.59
     collectively
    -0.58
    POSITIVE LOGITS
     himself
    1.83
     Himself
    1.24
     his
    1.13
     herself
    1.04
     personally
    0.92
     wife
    0.86
     solo
    0.83
    enegger
    0.80
    itone
    0.79
     resigned
    0.79
    Activations Density 6.330%

    No Known Activations