© 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 49152
    9. 8-RES_FS49152-JB
    11. 45707
    Prev
    Next
    INDEX
    Explanations

    phrases related to healthcare policies and immigration issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    87
    +0.20
    1.0%
    756
    +0.14
    0.7%
    86
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.20
    -0.01
    570
    +0.14
    0.02
    86
    +0.10
    0.04
    Negative Logits
     humour
    -0.90
     rumours
    -0.87
     enqu
    -0.86
     organisers
    -0.84
     organising
    -0.79
     colourful
    -0.76
     prank
    -0.74
     apologise
    -0.74
     realise
    -0.74
     learnt
    -0.72
    POSITIVE LOGITS
     ObamaCare
    1.16
     Obamacare
    1.15
     Medicare
    1.13
     Medicaid
    1.13
    amacare
    1.08
     CBO
    1.05
     uninsured
    1.02
     federally
    0.98
     HHS
    0.90
     subsidized
    0.90
    Activations Density 0.545%

    No Known Activations