© 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
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10263
    Prev
    Next
    INDEX
    Explanations

    phrases related to military personnel, operations, and events

    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
    288
    +0.15
    0.7%
    152
    +0.15
    0.7%
    146
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.15
    0.06
    288
    +0.15
    0.05
    146
    +0.10
    0.04
    Negative Logits
    pert
    -0.80
    uv
    -0.79
    Elsa
    -0.78
    bars
    -0.77
    theless
    -0.74
    alys
    -0.71
    flush
    -0.71
    à
    -0.68
     pert
    -0.67
    izers
    -0.66
    POSITIVE LOGITS
     Corps
    1.12
     Academy
    1.11
     Reserve
    1.10
     Veterans
    1.05
     Personnel
    1.04
     Veteran
    0.97
     SEAL
    0.95
     colonel
    0.93
     Command
    0.93
     brass
    0.92
    Activations Density 0.125%

    No Known Activations