© 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. 9846
    Prev
    Next
    INDEX
    Explanations

    information related to official records and legal proceedings

    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
    +0.15
    0.7%
    218
    +0.15
    0.7%
    230
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    230
    +0.15
    0.04
    268
    +0.15
    0.03
    575
    +0.10
    0.04
    Negative Logits
    Interstitial
    -0.96
    âĶģ
    -0.77
    liam
    -0.71
    ominated
    -0.68
    Solution
    -0.67
    odor
    -0.67
    agonist
    -0.67
    ãĥĦ
    -0.67
    culosis
    -0.65
    axis
    -0.65
    POSITIVE LOGITS
     unearthed
    1.16
     declass
    1.14
    archive
    1.09
     obtained
    1.08
     leaked
    1.04
     trove
    1.03
     detailing
    1.02
     records
    1.02
     documents
    1.00
     subpoen
    0.99
    Activations Density 0.242%

    No Known Activations