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

    Neuronpedia

    Jacobian LensNEW
    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 24576
    9. 8-RES_FS24576-JB
    11. 12987
    Prev
    Next
    INDEX
    Explanations

    words related to financial crimes and enforcement, such as piracy, fraud, evasion, corruption, laundering, and money

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.21
    1.0%
    481
    +0.13
    0.6%
    288
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.21
    0.06
    126
    +0.13
    0.04
    439
    +0.11
    0.04
    Negative Logits
    iman
    -0.77
    ube
    -0.76
    ixed
    -0.74
    ellar
    -0.73
    cise
    -0.68
    arine
    -0.66
    pret
    -0.66
    ŃĶ
    -0.64
    umm
    -0.64
     Bake
    -0.64
    POSITIVE LOGITS
     perpetrated
    1.12
     prevention
    1.08
     tactics
    1.05
     accusations
    1.03
     allegations
    0.99
     offences
    0.96
     accusation
    0.92
     prosecutions
    0.92
     crimes
    0.92
     plag
    0.92
    Activations Density 0.174%

    No Known Activations