© 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 24576
    9. 8-RES_FS24576-JB
    11. 9346
    Prev
    Next
    INDEX
    Explanations

    dates or time references

    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
    292
    +0.09
    0.4%
    123
    +0.09
    0.4%
    296
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    611
    +0.09
    0.03
    657
    +0.09
    0.03
    33
    +0.09
    0.03
    Negative Logits
     Cosponsors
    -0.85
     guessed
    -0.71
    ographical
    -0.69
     vou
    -0.67
    folk
    -0.65
    rices
    -0.65
    istas
    -0.64
    RED
    -0.64
    webkit
    -0.64
     nods
    -0.64
    POSITIVE LOGITS
     Ramadan
    0.84
    imester
    0.81
     gestation
    0.75
     hostilities
    0.72
     tenancy
    0.71
     adolescence
    0.70
     reckoning
    0.70
     occupancy
    0.70
     existence
    0.69
     adolesc
    0.69
    Activations Density 0.059%

    No Known Activations