© 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. 1426
    Prev
    Next
    INDEX
    Explanations

    government and economic related terms

    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
    480
    +0.50
    2.8%
    447
    +0.25
    1.4%
    481
    +0.20
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.50
    0.02
    288
    +0.25
    0.15
    481
    +0.20
    0.13
    Negative Logits
    $.
    -0.62
    ]+
    -0.59
    yond
    -0.56
    ãģĻ
    -0.53
    +.
    -0.53
    Translation
    -0.52
     Palest
    -0.52
    Els
    -0.50
     eternity
    -0.49
    âĶĢâĶĢâĶĢâĶĢ
    -0.48
    POSITIVE LOGITS
     has
    1.03
     reacted
    0.96
     took
    0.90
     appears
    0.88
     is
    0.88
     went
    0.88
     consists
    0.87
     may
    0.86
     had
    0.86
     arrived
    0.85
    Activations Density 1.654%

    No Known Activations