© 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. 23013
    Prev
    Next
    INDEX
    Explanations

    references to a governing or regulatory body, particularly in the context of commissions

    oai_token-act-pair · gpt-4o-miniTriggered 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
    746
    +0.11
    0.5%
    314
    +0.11
    0.5%
    149
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    644
    +0.11
    0.02
    146
    +0.11
    0.02
    746
    +0.10
    0.01
    Negative Logits
     prest
    -0.71
    vae
    -0.67
     Tibetan
    -0.62
    comes
    -0.62
     Kush
    -0.61
     flu
    -0.61
    imm
    -0.61
    hod
    -0.60
     Dharma
    -0.59
    Els
    -0.59
    POSITIVE LOGITS
    issions
    0.91
     commissioners
    0.91
    ing
    0.83
     commissions
    0.82
    eur
    0.80
     Commissioners
    0.78
    ission
    0.77
     commission
    0.77
    ional
    0.76
    ittee
    0.74
    Activations Density 0.020%

    No Known Activations