© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 3072
    9. 8-RES_FS3072-JB
    11. 126
    Prev
    Next
    INDEX
    Explanations

    activities or situations related to law enforcement and investigations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    218
    +0.14
    0.7%
    481
    +0.11
    0.5%
    511
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.14
    0.10
    1
    +0.11
    0.08
    218
    +0.10
    0.06
    Negative Logits
    minus
    -0.79
    wrap
    -0.77
    ģĸ
    -0.77
    urai
    -0.75
    hement
    -0.74
    phabet
    -0.73
    chart
    -0.73
    inion
    -0.73
    utterstock
    -0.73
     metaphor
    -0.73
    POSITIVE LOGITS
     suspicious
    1.32
     anomalies
    1.20
     abnormalities
    1.19
     suspected
    1.19
     irregularities
    1.16
     wrongdoing
    1.14
     discrepancies
    1.13
     anomal
    1.10
     culprit
    1.05
     disturbances
    1.04
    Activations Density 2.039%

    No Known Activations