Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 9942
    Prev
    Next
    INDEX
    Explanations

    mentions of crime-related incidents involving women, medical treatment, police reports, and server vulnerabilities

    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
    +0.48
    2.6%
    87
    +0.20
    1.1%
    266
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.48
    0.04
    266
    +0.20
    0.13
    720
    +0.15
    0.07
    Negative Logits
     rul
    -0.67
     conclud
    -0.64
    _.
    -0.61
    ãĥ´ãĤ¡
    -0.58
    External
    -0.57
     Vaugh
    -0.56
     incorpor
    -0.55
    REDACTED
    -0.54
    DK
    -0.54
    @@@@@@@@
    -0.53
    POSITIVE LOGITS
    their
    0.87
     their
    0.83
    utterstock
    0.79
     Their
    0.75
     THEIR
    0.73
     themselves
    0.72
    selves
    0.68
    Their
    0.67
    They
    0.66
     Shutterstock
    0.66
    Activations Density 1.892%

    No Known Activations