© 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. 323
    Prev
    Next
    INDEX
    Explanations

    descriptive scenes and settings in stories or articles

    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
    447
    +0.12
    0.5%
    709
    +0.11
    0.5%
    747
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    747
    +0.12
    0.02
    709
    +0.11
    0.02
    438
    +0.09
    0.02
    Negative Logits
    thood
    -0.96
    berman
    -0.80
    heit
    -0.80
    orem
    -0.79
    gements
    -0.79
    gemony
    -0.76
    sonian
    -0.74
    achev
    -0.74
    dp
    -0.74
    rition
    -0.74
    POSITIVE LOGITS
     scenes
    1.03
    Dialogue
    0.99
    Capture
    0.96
     Scenes
    0.95
     unfolding
    0.95
     involving
    0.93
     Scene
    0.90
     unfold
    0.83
     depicting
    0.81
    scene
    0.81
    Activations Density 0.073%

    No Known Activations