© 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 1536
    9. 8-RES_FS1536-JB
    11. 1080
    Prev
    Next
    INDEX
    Explanations

    phrases related to political and ideological beliefs

    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
    737
    +0.12
    0.5%
    98
    +0.11
    0.5%
    583
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    737
    +0.12
    0.13
    8
    +0.11
    0.13
    762
    +0.11
    0.17
    Negative Logits
     Reported
    -0.86
    untled
    -0.86
    umably
    -0.82
    recent
    -0.82
    Previously
    -0.77
    USS
    -0.77
    mentioned
    -0.76
    oof
    -0.71
     extensively
    -0.70
     Presumably
    -0.70
    POSITIVE LOGITS
     righteousness
    1.21
     destiny
    1.11
     dignity
    1.07
     greatness
    1.05
     patri
    0.96
     salvation
    0.96
     courage
    0.95
     tomorrow
    0.94
     sacrifice
    0.93
     prosperity
    0.93
    Activations Density 6.618%

    No Known Activations