© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 10141
    Prev
    Next
    INDEX
    Explanations

    phrases related to unity and shared humanity

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    +0.27
    1.4%
    756
    +0.14
    0.7%
    283
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.27
    -0.02
    87
    +0.14
    -0.04
    283
    +0.12
    0.03
    Negative Logits
    DragonMagazine
    -0.71
    awaited
    -0.63
    etsk
    -0.61
     Missing
    -0.58
     Shutdown
    -0.57
    umbai
    -0.57
    Draft
    -0.57
     Recovery
    -0.56
     Solitaire
    -0.55
     Launcher
    -0.55
    POSITIVE LOGITS
     ideologies
    0.81
     societies
    0.79
     society
    0.76
     abstract
    0.73
     perce
    0.73
     worldview
    0.73
     oneself
    0.73
     civilizations
    0.72
     capitalism
    0.72
     humankind
    0.72
    Activations Density 1.825%

    No Known Activations