© 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. 9761
    Prev
    Next
    INDEX
    Explanations

    mentions of specific roles or identities in reference to people

    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
    715
    +0.13
    0.6%
    567
    +0.13
    0.6%
    302
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.13
    0.05
    660
    +0.13
    0.05
    698
    +0.12
    0.04
    Negative Logits
    ories
    -0.88
    itars
    -0.79
    ernels
    -0.77
     divisions
    -0.76
    metadata
    -0.75
    irens
    -0.75
    ources
    -0.73
     Horses
    -0.73
    acements
    -0.72
    ippers
    -0.72
    POSITIVE LOGITS
     who
    1.00
     chooses
    1.00
     learns
    0.98
     knows
    0.97
     understands
    0.91
     decides
    0.84
    who
    0.84
     realizes
    0.84
     spends
    0.81
     sees
    0.80
    Activations Density 0.219%

    No Known Activations